Upload
others
View
7
Download
0
Embed Size (px)
Citation preview
A Cross-Cultural Evaluation of
Individualism and Collectivism on
Communication Strategies and Social
Identity on Online Social Media
A thesis submitted in partial fulfilment of the
requirements of Nottingham Trent University
for the degree of Doctor of Philosophy
By
Moon Moon Halder
Nottingham Trent University
School of Social Sciences
October 2016
This work is the intellectual property of the author. You
may copy up to 5% of this work for private study, or
personal, non-commercial research. Any re-use of the
information contained within this document should be
fully referenced, quoting the author, title, university,
degree level and pagination. Queries or requests for any
other use, or if a more substantial copy is required,
should be directed in the owner(s) of the Intellectual
Property Rights.
i
Dedication
This thesis is dedicated to:
My beloved father “Baba”, you are always in my memory and my mother “Maa”, Thank you
for always being there. Thank you for your blessings.
My uncle “Sona Jethu” – Thank you for believing in me.
My brother “Rahul”, sister “Priyanka”, my nieces “Ishika”, “Toshumi” and nephew “Deep” -
Thank you all for being so encouraging and pushing me to strive harder.
And above all,
My dear husband “Akil” – You have been the pillar of strength for me. Thank you very much
for believing in me. Without your support and patience my research would not have reached
its completion. Thank you for being there. Thank you for all your support and encouragement
which has helped me to persevere through thick and thin times. This achievement is as much
yours as it is mine.
ii
Acknowledgements
This thesis would not have been possible without the support and encouragement of many
people. While it would be difficult to list the names of every staff, faculty, student, family
member and friend who has helped in this process and kept the thesis going. I would like to
extend my deepest thanks and gratitude to each one of who has helped in every possible way
in this process.
I would especially like to extend my thanks and gratitude to my supervisors’ Dr Jens
Binder (Director of Studies), Dr James Stiller and Dr Mick Gregson who have all been really
supportive and encouraging throughout this process. I am grateful for their input which has
helped shape this thesis. Thank you, Dr James Stiller and Dr Jens Binder for providing the
time and support, whenever I needed, which has helped elevate the quality of my work. I will
always be grateful to you both for being there. I would not be presenting this work if it was
not for the encouragement and mentoring of Dr Jens Binder and Dr James Stiller.
I would also like to thank staff and students at St. Thomas School (India), Universitas
Gadjah Mada (Indonesia), Universitas Ahmad Dilan (Indonesia), Indonesian Islamic
University (Indonesia) and especially to MS Wendy Leyland and Mrs Indrayanti for helping
me with my data collection.
Above all, I would like to thank my family for their love and support they have provided
throughout and for having faith in my work.
I would also like to Thank my uncle (Dr Tapon Halder) and my aunty (“Sona Maa”) for having
faith in me.
My acknowledgements would not be complete without offering my gratitude to my dear
husband (“Akil”) who has been my pillar of strength throughout this process and being my
guiding path.
iii
Abstract
The increase of the use of social network sites (SNS) has given the opportunity for
members from all cultures to maintain existing and establish new social connections and also
create an online social identity for themselves. Such online communication has now become
part of our daily lives, where members from individualist (independent) and collectivist
(interdependent) cultures interact with each other. While past research has been successful in
understanding the causal role of culture in an offline context which has affects social identity
and psychological understanding and decisions members make, the evaluation of how culture
effects human behaviour and their social identity in an online environment is limited.
This research aims to address this gap in literature by acknowledging the concept of Social
Identity Model of Deindividuation Effect (SIDE) which posits that the level of anonymity in
an online environment helps members to depersonalise, thereby helping in social identification
of the self and facilitating online relationships and communication. This research particularly
evaluates how cultural differences effects online communication, the decisions members
making in various psychological outcomes and how if effects their social identities.
This will be achieved by a cross-cultural among Facebook members from individualist
country (UK) and collectivist country (India and Indonesia) through a survey and priming
experiments. While cultural orientation has been used as an independent variable in the study,
psychological variables like online social identity, online self-enhancement, perceived online
social support, online trust and online privacy concern has been used as the dependent
variables in the study.
The research used a quantitative method approach which was divided into four studies.
Study 1 (N = 150) and Study 2 (N = 432) consisted of the online study involving Facebook
responses and Study 3 (N = 71) and Study 4 (N = 407) consisted of the priming experiments
where cultural orientation was manipulated to check participant responses on various
psychological outcome measures.
iv
The results revealed that differences in participant responses existed among both cultures.
Priming members with consistent cultural self-construal strengthened their scores on their
cultural orientation and when members were primed with inconsistent cultural self-construal,
it weakened their scores on their cultural orientation. The causal role of culture was established
in some of the outcome measures that will be discussed further.
The results highlight the importance of understanding and acknowledging cultural
differences of members using SNSs. This is not just crucial for health professionals trying to
incorporate SNSs to implement care to patients, the results also highlight the important
responsibility for web developers and network providers fighting privacy issues, online
bullying. Additionally, the results are also crucial to social researchers as they try to
understand online behaviour and to the members of SNSs itself to help prevent online tension.
v
Table of Contents
Acknowledgement iv
Abstract v
Table of Contents iiv
List of Tables xv
List of Figures xvii
Chapter One: Introduction
1.1. Overview……………………………………………………….…………...…………… 1
1.2. Problem Statement……………………………………………....................................... 2
1.3. Significance of the Study…………………………………………………...…………... 2
1.4. Outline of Methodology……………………………………………………...…………. 4
1.5. Structure of the Study (Overview)…………………………………………...………… 4
Chapter Two: Theoretical Framework
2.1. Literature review (Theoretical base) …………………………………….…………….........7
2.1.1. Culture…………………………………………………………………………….... 7
2.1.2. Culture and Self-construal …………………………………….………………….. 9
2.1.3. The Social Identity Theory ………………………………………………………. 10
2.1.4. Self-categorization Theory ………………………………………………………. 12
2.1.5. Social Identity Model of Deindividuation (SIDE) …………………………….… 14
vi
2.2. Literature Review (Psychological variables used in the study) ………………………….. 16
2.2.1. Intercultural Importance of Social Media ……………….……………………… 16
2.2.2. Online Group Identity ………………………………………...…………………. 20
2.2.3. Online Self-enhancement ………………………………………………………… 25
2.2.4. Online Perceived Social Support ………………………………………………… 30
2.2.5. Online Trust …………………………………………………………………...…. 34
2.2.6. Online Privacy Concern ……………………………………………...………….. 38
2.3. Summary …………………………………………………………………………...……... 42
2.4. Research Aims and Hypotheses ………………………………………………………….. 44
2.5. Brief Outline of Research Overview ………………………………………………….….. 45
Chapter Three: Methodology
3.1. Introduction ………………………………………………….…………………………… 48
3.2. Research Paradigm and Outline of Design ………………………...……………………. 48
3.3. Research Population ………………………………………………………....………........ 48
3.4. Research Considerations …………………………………………………………...…….. 50
3.4.1. Cultural consideration …………………………………………...…………......... 50
3.4.2. Language consideration ………………………………………………...……….. 50
3.4.2.1. Survey Translation Procedure ……………………………………………... 50
3.4.2.1.1. Back Translation Procedure for research instrument in India ………52
3.4.2.1.2. Back Translation Procedure for Indonesia ……………………………53
3.4.3. Ethical considerations …………………………………………………………..... 54
vii
3.5. Overview of Study Design …………………………………………………...…………… 55
3.6. Summary ……………………………………………………………………………..…… 56
Chapter Four: Study 1: Survey (Operationalization)
4.1. Rationale for this study ………………………………………………………………….... 58
4.2. Method ……………………………………………………………………………………. 59
4.2.1. Participants ………………………………………………………………………. 59
4.2.2. Procedure ………………………………………………………………………… 59
4.2.3. Materials & Description of items ……………………………………….………. 59
4.2.3.1. Cultural self-construal scale ………………………………………...……… 61
4.2.3.2. Facebook Activity Scale ………………………………………..…………… 61
4.2.3.3. Online Group Identification Scale ……………………………..………….. 62
4.2.3.4. Online Self-Enhancement Scale ……………………………..…………….. 62
4.2.3.5. Perceived Online Social Support Scale ……………………………………. 63
4.2.3.6. Online Trust Scale ………………………………………………………….. 64
4.2.3.7. Online privacy scale ………………………………………………………… 64
4.3. Results and Analysis ……………………………………………………………………… 64
4.3.1. Descriptive statistics ………………………………………………..…………….. 65
4.3.2. Data screening and Preliminary analysis ………………………..……………… 65
4.3.3. Sample characteristics ……………………………………………..…………….. 65
4.3.4. Analysis ……………………………………………………………..…………….. 76
4.4. Brief summary of Study 1 …………………………………………………...……………. 77
viii
4.5. Changes made after completion of Study 1 ……………………………………………… 78
Chapter Five: Study 2: Survey (Facebook Responses)
5.1. Rationale for this study …………………………………………………………………… 80
5.2. Method …………………………………………………………………………………….. 82
5.2.1. Participants …………………………………………………………...………….. 82
5.2.2. Procedure …………………………………………………………..…………….. 83
5.2.3. Materials & Item description ……………………………………...…………...... 83
5.2.3.1. Cultural Self-Construal Scale …………………………..………………. 83
5.2.3.2. Online Facebook Activity Scale ………………………..………………... 83
5.2.3.3. Online Group Identification Scale ………………..…………………….. 84
5.2.3.4. Online Self-Enhancement Scale ……………………………...…...…….. 84
5.2.3.5. Perceived Online Social Support Scale ……………………...………….. 85
5.2.3.6. Online Trust scale …………………………………………..…………… 85
5.2.3.7. Online Privacy Concern Scale …………………………...……………… 85
5.3. Results ……………………………………………………………………..………………. 86
5.3.1. Descriptive statistics ……………………………………………...………………. 87
5.3.2. Data screening and Preliminary analysis ………………………………………. 87
5.3.3. Sample characteristics …………………………………………………………… 88
5.3.4. Investigating the relationship between variables ………………………….……. 93
5.3.5. Check for Moderation ………………………………………………………….... 95
5.3.6. Mediation Analysis ………………………………………………………...……. 96
ix
5.3.7. Modelling of Multivariate Relationships ………………………………………. 98
5.3.7.1. Preparatory analyses ……………………………………………………...... 98
5.3.7.2. Reliability Measure ……………………………………………………...…. 99
5.3.8. Online Facebook Activity Scale ………………………………………………… 101
5.3.9. Online Cultural Self-Construal Scale ………………………………………….. 102
5.3.10. Online Group identity …………………………………………………………. 103
5.3.11. Online Self enhancement ……………………………………………………… 103
5.3.12. Perceived Online Social support ……………………………………………… 104
5.3.13. Online Trust ………………………………………………………………...…. 105
5.3.14. Online Privacy Concern …………………………………………………...….. 105
5.3.15. SEM Analysis ………………………………………………………….………. 107
5.3.16. Model specification …………………………………………………….……… 107
5.3.17. Model Estimation ……………………………………………………….……... 108
5.3.17.1. The chi square (X2) ………………………………………………….……. 108
5.3.17.2. Goodness-of-Fit Index (GFI) ……………………………………….……. 108
5.3.17.3. Root Mean Square Residual Index (RMR) ………………………….….. 109
5.3.17.4. Normal Fit Index (NFI) and Comparative Fit Index (CFI) ………...…... 109
5.3.17.5. Root Mean Square Error of Approximation (RAMSEA) …………..….. 109
5.3.18. Model Testing ……………………………………………………………….…. 109
5.3.19. Model Modification ………………………………………………………….… 109
5.3.20. Re-specification ………………………………………………………………... 109
5.3.21. Process ………………………………………………………………….……… 110
5.3.21.1. Online Group Identity …………………………………………….……... 111
x
5.3.21.2. Online Self-Enhancement ……………………………………................... 117
5.3.21.3. Perceived Online Social Support ……..………………………….............. 126
5.3.21.4. Online Trust ……..………………………….............................................. 131
5.3.21.5. Online Privacy Concern ……..……………............................................... 137
5.9. Summary of Results …………………………………………………………………….. 143
Chapter Six: Study 3: Experiment (Operationalization)
6.1. Rationale for this Study …………………………………………………………………. 146
6.2. Method …………………………………………………………………………………… 147
6.2.1. Participants ………………………………………………………………………… 147
6.2.2. Procedure …………………………………………………………………………... 147
6.2.3. Material & Description of items …………………………………………..……….. 147
6.3. Results and Analysis …………………………………………………………….………. 151
6.3.1. Data screening and Preliminary analysis …………………………………..……... 151
6.4. Summary ………………………………………………………………………….……... 156
6.5. Changes made after completion of Study 3 ……………………………………...……... 156
Chapter Seven: Study 4: Experiment (Priming & Facebook responses)
7.1. Rationale for this study …………………………………………………………..……… 159
7.2. Method …………………………………………………………………………….……... 162
7.2.1. Participants ………………………………………………………………….……... 162
7.2.2. Procedure …………………………………………………………………….…….. 162
xi
7.2.3. Materials &Description of items ……………………………………………..……. 162
7.3. Results and Analysis ……………………………………………………………….……. 166
7.3.1. Data screening and Preliminary analysis ……………………………………..…... 166
7.3.2. Sample characteristics …………………………………………………………..…. 167
7.3.3. Twenty Statements Test (TST) of Self-construal ………………………………..... 169
7.3.4. Online Group Identity ………………………………………………………..……. 171
7.3.5. Online Self-Enhancement ……………………………………………………..…… 173
7.3.6. Perceived Online Social Support ……………………………………………...…… 175
7.3.7. Online Trust …………………………………………………………………….….. 176
7.3.8. Online Privacy Concern ………………………………………………………...…. 177
7.3.9. Scenario Task ……………………………………………………………...……….. 178
7.3.9.1. Friend request from family member…………………………………...…….. 178
7.3.9.2. Friend request from unknown person ……………………………………….. 179
7.3.9.3. Update privacy settings ……………………………………………………….. 180
7.3.9.4. Join protest for a noble/ social cause ……………………………………….… 181
7.3.9.5. Acknowledge success of others ……………………………………………….. 182
7.2.9.6. Join social group having personal importance …………………….………… 182
7.4 Summary ………………………………………………………………………...……….. 184
7.5. Comparison of Results …………………………………………………………………... 188
Chapter Eight: Discussion
8.1. Overview of Findings ……………………………………………………………………. 192
xii
8.2. Limitations ……………………………………………………………………...………. 201
8.3. Contributions of the Study ……………………………………………………………… 202
8.4. Implications and Future Research ……………………………………………………… 203
8.5. Conclusion………………………………………………………………………………... 204
xiii
List of Tables
3.1. An Illustration of the Study Design for the Thesis………………….………………….. 56
4.1. Demonstrates the descriptive for both UK and Indonesian sample……………..…….. 66
4.2. Demonstrates Ethnicity and Work characteristics participants in UK and Indonesia .67
4.3. Demonstrates FB use per weekday & per weekend in UK & Indonesia (in hours) ...... 69
4.4. Illustrates Frequency of Facebook use among UK & Indonesian sample …………..... 70
4.5. Reliability scores of the variables used in the Survey ………………………………….. 71
4.6. Illustrates correlations between variables (unstandardized items) …………………….. 73
4.7. Illustrating the Standardized Regression Coefficients predicting FB Activities .......... 75
5.1. Reliability scores of the variables used in Study 2……………………………………… 86
5.2. Illustrates Descriptive statistics of participants in Study 2……………………………. 88
5.3. Demonstrates Ethnicity and Work characteristics of participants in Study 2 …….…. 89
5.4. Demonstrates Facebook use during weekday and weekends in UK and Indonesia …. 90
5.5. Demonstrates Facebook use among UK and Indonesian respondents in Study 2 ..….. 91
5.6. Demonstrates inter-correlation of the variables in Study 2 ………………………….... 93
5.7. Demonstrates the Indirect Effect of Cultural Orientation on Outcome Variables…... 96
5.8. Demonstrates the original items and the amended number of items (N) for each
variable, their Cronbach alpha (α), the Mean and SD of the amended variables……….. 100
5.9. Demonstrates variables with their factor loadings, percentage of variance, number of
items (N) and their Cronbach alpha score [a]……………………………………………… 106
xiv
5.10. Illustrates the Model Fit Indices, the Mediation Effect (Indirect Effect) & the
Confidence Intervals for all the models for Online Group Identity …………………….... 115
5.11. Illustrates the Model Fit Indices, the Mediation Effect (Indirect Effect) & the
Confidence Intervals for all the models for Online Self-Enhancement ………………….. 124
5.12. Illustrates the Model Fit Indices, the Mediation Effect (Indirect Effect) & the
Confidence Intervals for all the models for Perceived Online Social Support ………….. 130
5.13. Illustrates the Model Fit Indices, the Mediation Effect (Indirect Effect) & the
Confidence Intervals for all the models for Online Trust ………………………………… 136
5.14. Illustrates the Model Fit Indices, the Mediation Effect (Indirect Effect) & the
Confidence Intervals for all the models for Online Privacy Concern ……………………. 142
6.1. Illustrates the Descriptive statistics for participants in Study 3 ……………………... 152
6.2. Ethnicity and work characteristics of the participants in Study 3 …………………... 152
6.3. Demonstrates the Inter-Correlation of the variables in Study 3 …………………….. 154
7.1. Sample Characteristics, Study 4 ……………………………………………………….. 167
7.2. Ethnicity and work characteristics of the participants in Study 4 …………………... 168
7.3. Illustrates the results of the hypotheses in Study 4 and its outcomes ……………….. 184
xv
List of Figures
2.1. Visual Representation of the Main Constructs used in the Study …………………… 43
2.2. Visual Illustration of Research Overview ………………………………………………47
3.1. An Illustration of the Back Translation Process followed in India & Indonesia ……. 54
5.1. Operational Theoretical Model showing all the variables and mediators ……………. 81
5.2. Demonstrates the Basic Model that illustrates the relationship between Cultural
Orientation, Facebook Engagement and Psychological Outcome Variables ………………82
5.3. Models 1 to 4 Illustrates Standardized Path Coefficients of the Mediating Effect on
Online Group Identity using Collectivism as the predictor variable …………………….. 113
5.4. Models 5 to 8 Illustrates Standardized Path Coefficients of the Mediating Effect on
Online Group Identity using Individualism as the predictor variable …………..………. 114
5.6. Models 1 to 4 Illustrates Standardized Path Coefficients of the Mediating Effect on
Online Self-Enhancement using Collectivism as the predictor variable on Collectivist
Attributes …………………………………………………………………………………….. 119
5.7. Models 5 to 8 Illustrates Standardized Path Coefficients of the Mediating Effect on
Online Self-Enhancement using Collectivism as the predictor variable on Individualist
Attributes …………………………………………………………………………………….. 120
5.8. Models 9 to 12 Illustrates Standardized Path Coefficients of the Mediating Effect on
Online Self-Enhancement using Individualism as the predictor variable on Collectivist
Attributes …………………………………………………………………………………….. 121
xvi
5.9. Models 13 to 16 Illustrates Standardized Path Coefficients of the Mediating Effect on
Online Self-Enhancement using Individualism as the predictor variable on Individualist
Attributes …………………………………………………………………………………….. 122
5.12. Models 1 to 4 Illustrates Standardized Path Coefficients of the Mediating Effect on
Perceived Online Social Support using Collectivism as the predictor variable …………. 127
5.13. Models 5 to 8 Illustrates Standardized Path Coefficients of the Mediating Effect on
Perceived Online Social Support using Individualism as the predictor variable …………. 128
5.15. Models 1 to 4 Illustrates Standardized Path Coefficients of the Mediating Effect on
Online Trust with Collectivism as the predictor variable …………………………………. 133
5.16. Models 5 to 8 Illustrates Standardized Path Coefficients of the Mediating Effect on
Online Trust with Individualism as the predictor variable ……………………………….. 134
5.17. Model 1 to 4 Illustrates Standardized Path Coefficients of the Mediating Effect on
Online Privacy Concern using Collectivism as the predictor variable …………………… 138
5.18. Model 5 to 8 Illustrates Standardized Path Coefficients of the Mediating Effect on
Online Privacy Concern using Individualism as the predictor variable …………………. 139
6.1. Graphical illustration of the interaction effect of country and priming on TST scores in
UK and India …………………………………………………………………………………. 156
7.1. Graphical interpretation of the interaction effect on the TST responses …………….. 171
7.2. Graphical interpretation of the interaction effect on Group Identity responses ……. 172
7.3. Graphical interpretation of the interaction effect of pre and post priming ratings on Self-
Enhancement on Collectivism attributes …………………………………………………… 174
7.4. Graphical interpretation of the interaction effect of pre and post priming ratings on Self-
Enhancement on Individualist attributes …………………………………………………… 175
xvii
7.5. Graphical illustration of the interaction effect of country and priming on the scores on
Online Trust in UK and Indonesia ………………………………………………………….. 177
7.6. Graphical illustration of the interaction effect of country and priming on the scores on
Online Privacy Concern in UK and Indonesia ……………………………………………... 178
7.7. Graphical illustration of the interaction effect of country and priming (Scenario 2) . 180
7.8. Graphical illustration of the interaction effect of country and priming (Scenario 3) ……….. 181
7.9. Graphical illustration of the interaction effect of country and priming (Scenario 4) ………... 182
7.10. Graphical illustration of the interaction effect of country and priming (Scenario 6) ………. 183
1
Chapter One
Introduction
1.1 Overview
The advent of the internet has transformed the way people communicate. People now have
progressed from face-to-face communication to communicating over the internet using Social
Networking Sites (SNS) like Facebook, Twitter, MySpace, and YouTube, which has become part
of our daily lives. The increased use of the internet has revolutionised the way we interact with
each other, not just on an interpersonal level, but also when it comes to communicating with our
social groups (Raacke & Bonds-Raacke, 2008). While the traditional form of a social network
consists of our families, close friends and colleagues, the use of SNS has given the opportunity to
not just maintain our old ties but also add new ones (Burke, Marlow, & Lento, 2010; Ellison,
Steinfield, & Lampe, 2006; Ellison, Steinfield, & Lampe, 2007). SNS have also been able to help
in maintaining long distance relationships (Tosun, 2012) and have also been seen to have
psychological benefits such as increased perceptions of social capital (Ellison et al., 2007; Lampe,
Ellison, & Steinfield, 2008). However excessive use of SNS has also been linked to addiction
(Kuss & Griffiths, 2011; Kuss, Griffiths, & Binder, 2013) and other privacy risks. Although people
are aware of the potential risks associated with using SNS, they are still happy to disclose
information due to the benefits that come with it (Debatin et al., 2009).
Use of SNS is not a western phenomenon but is a part of daily lives globally due to the
accessibility and the availability of the internet. As of September 2016, Facebook alone has 1.71
billion monthly active users, the photo sharing application Instagram had over 500 million monthly
active accounts and blogging sites like Tumblr had more than 555 million active bloggers on their
sites (Statista, 2016). On Facebook alone the average number of friends is 338 (brandwatch, 2016).
Due to the diverse population of its members, an attempt to understand online behaviours on SNS
cannot be done without acknowledging the role of culture. For example, Facebook itself offers its
site in over 70 languages, Twitter offers its site in over 21 languages and LinkedIn offers its site
in over 17 languages (Statista, 2016) which highlight the cultural diversity on SNS.
2
From a psychological perspective, the cultural variability which shapes one’s self-construal
(Triandis, 1989) would have an impact on how SNS members perceive and generate information
online which also has an impact on further online behaviours. While there has been an immense
body of past research that has been conducted to understand cultural influences on the self and
behaviour (Bochner, 1994; Gardner, Gabriel, & Lee, 1999; Gudykunst et al., 1996; Hui & Triandis,
1986; Markus & Kitayama, 1991; Nisbett et al., 2001; Triandis, 1989), which helps to enhance our
knowledge on interpersonal communication. There exist differences in behaviour and attitude of
individuals even when they come from the same nation (Hofstede & Hofstede, 2001) so it could
be speculated that when members from different cultures interact on SNS, user behaviour would
also vary. Given the diversity of users on SNS it is therefore important to consider the causal role
of culture on user behaviour if researchers want to understand online behaviour. Therefore, this
research aims to conduct a cross-cultural research with natives from individualist and collectivist
cultures (see Hofstede, 1980) to address the diversity on SNS.
1.2. Problem statement
While past research has mainly focused on understanding human behaviour in an offline
context, the study of online behaviour is still limited. Thus, the present research attempts to address
this research gap in the empirical literature. Furthermore, the growing use of SNS among diverse
populations from around the globe, calls for an evaluation of the causal role of culture in an online
context. Additionally, the use of validated self-report scales in cross-cultural research, which have
been developed mostly on the basis of western attitudes and behaviour is also questionable.
1.3. Significance of the study
The goal of the present research was to investigate the cultural impact on communication
behaviour and social identities among individuals from different cultural backgrounds in an online
environment. Being able to study the differences in behaviour in different cultural context would
allow drawing stronger inferences from the data collected. The distinction between cultural self
was made by gauging one’s cultural self-construal by measuring their level of individualism and
3
collectivism (Hofstede, 1980). Although it has been argued that the goals of being autonomous
and belongingness are shared universally by all humans, as such members from both western and
eastern cultures can have both kinds of self-construal i.e., individualism (independent) and
collectivism (interdependent) self-construal (Brewer & Gardner, 1996; Cross, 1995; Trafimow,
Triandis, & Goto, 1991; Triandis, McCusker, & Hui, 1990) which is displayed based on one’s
goals and current situation an individual is in, however as the general cultural dimensions are
considered to be between levels of individualism and levels of collectivism, for this research
purpose the western population is considered as individualist country (as their members generally
would score higher on levels of individualism) and eastern population is considered as collectivist
country (as their members would generally score higher in levels of collectivism).
It was important to select psychological variables that could help evaluate user behaviour online
under several constructs in order to study online behaviour in-depth which constitutes a research
gap that this study aims to address. Next to cultural orientation, this research focusses on a range
of psychological variables and processes that have been identified as relevant in the online domain.
Due to the level of anonymity that exists during online communication and the lack of presence of
physical cues, the sustainability of such online communication depends on how users identify with
the others in the group (online group identity), which helps them to develop an online trusting
relationship (online trust), further motivating users to disclose information (online privacy
concern) in spite of potential risks due to the benefits that they get from such interaction (perceived
online social support and online self-enhancement). Hence perceived online social support, online
trust, online privacy concern, online group identity and online self-enhancement are the main
psychological outcome variables in the research.
This research aims to find differences in user activities and behaviour influenced by culture. It
has been envisaged that such differences in behaviours can be achieved using online survey and
priming experiment (Gardner, Gabriel, & Lee, 1999). The findings from this research will help not
just social scientist but web developers, medical practitioners and even academics and researchers.
4
1.4. Outline of Methodology
Keeping cultural orientation (IV) constant the effect of user behaviour was evaluated on the
various psychological variables (DV). An attempt was made to evaluate the differences in
behaviour through the use of various psychological variables along with differences in user
activities in an online environment, in particular on SNS like Facebook, use of survey techniques
was a good choice. Surveys, however, lead mostly to correlational evidence and are not a good
instrument to look at the causal role of cultural orientation on user behaviour. Experiments, on the
other hand, were the obvious method of choice when it comes to causal processes. Therefore, a
combination of methods was proposed which included an online survey which aimed to capture
user network and user behaviour on Facebook. The experiment proposed were a set of priming
tasks that was used on the participants and was used to show the causal role of culture on
individuals. The below section will now outline the structure of the thesis.
1.5. Structure of the Thesis (Overview)
The thesis consists of eight chapters and is organized as follows.
• Chapter Two: Literature Review
Literature and theoretical review of previous studies that have been conducted. This chapter has
several sub-sections which reviews past studies under each sub-section.
This sub-section reviews the theoretical basis of the research and consists of the below sub-
sections:
Culture; It evaluates differences in cultural behaviour in an online environment. It aims to provide
an overview how us of SNS is used globally.
Culture and self-construal: This chapter aims to outline how cultural norms and behaviour helps
in the development of self-construal. It highlights how different cultural dimensions.
Social Identity Theory: This chapter evaluates the importance of Social Identity Theory. It
highlights the importance of social identity in an online context.
5
Self-categorization Theory: This chapter outlines the importance of SCT in not just an intergroup
level but also at an intragroup level.
Social Identity Model of Deindividuation: This chapter outlines how the model of Social Identity
Theory is applied in an online context.
Facebook Use: This chapter justifies why Facebook has been used as the platform for my study. It
highlights how and why members around the globe use SNS like Facebook.
The below sub-sections review past literatures on the various psychological variables that has
been used as DV in the thesis.
Online Group Identity: It outlines the importance of social identity on SNS across cultures.
Online Self Enhancement: It outlines how the use of SNS has an impact on the level of self-
enhancement across cultures.
Perceived Online Social Support: It outlines how the use of SNS can help members with social
support across cultures.
Online Trust: It highlights the importance of trust on SNS.
Online Privacy Concern: It highlights how individuals across cultures behave due to privacy
concern in an online environment.
• Chapter Three: Methodology
This chapter outlines the method used in the study and discusses the rationale for adopting such
methods. It also discusses the various challenges that were faced during the cross-collection data
collection process.
• Chapter Four: Study 1: Survey (Operationalization)
Study 1 is used as a development study to refine study 2. It also consists of the results and its
analysis.
• Chapter Five: Study 2: Survey (Facebook Responses)
Study 2 was a refined version of Study 1 which consisted of an online survey. It also consists the
results and analysis.
6
• Chapter Six: Study 3: Experiment (Operationalization)
Study 3 consisted of the priming experiment and was also used as a study to refine the final version
of the experiment. Results and analysis for this study are also discussed in this chapter.
• Chapter Seven: Study 4: Experiment (Priming & Facebook Responses)
Study 4 was the refined and final version of the experimental study. This chapter also discusses
the results and its findings.
• Chapter Eight: Discussion
This chapter provides overview of findings in light of past research. It sheds light on the limitations
of the study, outlines the contributions of the study, discusses the implications and future research
along then finally provides a conclusion.
• References
• Appendices
7
Chapter Two
Literature Review
This section will outline literature review on the various constructs used in the thesis and the
chapter is divided into two sub-sections. Sub-section 2.1. is a literature review conducted that has
helped to form a theoretical base for this thesis and Section 2.2. will discuss past studies conducted
on the various psychological constructs that has been used as dependent variables (DVs) in the
thesis. Section 2.3. will provide a brief summary of this chapter followed by Section 2.4. which
consists of the Research aims and hypotheses and finally Section 2.5. will provide the Research
Overview.
2.1. Literature review (Theoretical base)
This section will discuss past literature review to help justify the theoretical base for this thesis
which will be used to evaluate differences in cross-cultural behaviour in an online environment.
2.1.1. Culture
Traditionally cross-cultural research has mainly focused on evaluating the role of culture by
looking at it from an offline context. For example, while the use of internet for communication
purpose is on the rise in present times, cross-cultural studies has been conducted way before this
period (Hofstede, 1980; Markus & Kitayama, 1991). However, the increased use of the internet
has fundamentally changed the way we communicate, from face-to-face communication to
communicating over the internet. With the recent increase in the use of Social Network Sites (SNS)
like Facebook, Twitter, My Space, YouTube, understanding cultural influence has become more
of a focus of interest for social researchers which has its benefits (Ellison, 2007; Ji et al., 2010;
Karl, Peluchette, & Schlaegel, 2010), revolutionizing the way we interact with each other, not just
on an interpersonal level, but also when it comes to communicating with our social groups (Raacke
& Bonds-Raacke, 2008) or even maintaining long distance relationships (Tosun, 2012). The use
8
of SNS has even been linked to addiction (Kuss & Griffiths, 2011; Kuss et al., 2013) also
highlighting the problematic nature of such use affecting behaviour of individuals.
This research focuses on understanding the communication behaviour on social media,
acknowledging the level of anonymity which helps members in the development of self-construal
and social identity online. SNS, in particular Facebook (www.facebook.com) will be used as a
platform to access research goals as it has access to a global sample of users from a wide range of
demographic backgrounds and is the biggest internet based company (brandwatch, 2016).
SNS function on the basis of verifying the level of anonymity and when members are able to
evaluate this, they are then able to connect to social groups. This shifts the focus from an individual
self to a group self thereby helping to change the perception of self and others from an interpersonal
level to a group level (see The Social Identity Model of Deindividuation - SIDE; Spears et al.,
2002). To the extent that online social media allow for anonymity, where members are at least able
to select and categorize their social groups, it will help members to communicate freely. Members
on this platform are able to create an identity for themselves (Krämer & Winter, 2008) and through
the profiles they create they are able to create a public image that they want others to perceive for
themselves (Lampe, Ellison, & Steinfield, 2007) which help members create editable an online
identity for themselves where they are able to express themselves without any restrictions.
However, such communication strategies will have cultural influence which will have an impact
on the behaviour of its members further stressing that the casual role of culture cannot be ignored
while evaluating the behaviour of members in an online environment. The below sections will
discuss the role of culture in the development of one’s self-construal “Culture and Self-Construal”,
followed by a discussion on the theoretical concept of Social Identity Theory, Social
Categorization Theory and Social Identity Model of Deindividuation (SIDE) which help shed light
the theoretical concepts which helped to form the base for this research. Further literature review
will be conducted on the use of SNS in general. Finally, the last section in this chapter will discuss
the various DVs that has been used in this thesis in light of past research.
9
2.1.2. Culture and Self-construal
The “self” is the mediating factor between culture and individual behaviour (Triandis, 1989).
The cultural norms and values that people follow with time becomes part of one’s behaviour which
help to shape and develop one’s self-construal. The self-construal of individual is a representation
of their cultural self and highlights the individual’s inner ability, characteristic, attribute or goal
which is influenced by the culture that individual might come from. For example, the normative
imperative of an individual from an individualist country possesses an independent self-construal
and thus encourage independence from others and being able to stand out and be able to express
their inner traits to others. In contrast, individuals coming from collectivist countries has
interdependent self-construal and where the self is connected to others (Markus & Kitayama, 1991;
Markus et al., 1998; Triandis, 1989). Collectivist members do have the concept of their inner self
but such inner attributes do not guide their decisions or behaviours in the same way as individuals
from individualist countries. Instead members from collectivist cultures emphasizes the
maintenance of group harmony which helps their construal becomes more meaningful and
complete. Members of the group are considered as the part of the self and the self is considered as
part of others. Their behaviours and actions are governed by what others perceive about them and
what they are required to do in order to maintain harmonious relationships. Collective existence
with their group members is a part of their self. The main differences between the two-types of
self-construal is how the self in relative to others in the social environment. While it is important
not to generalise such cultural dimensions to the whole population in Eastern and Western
countries but it will be fair to say that in general people in Western countries possess independent
concept of the self and a majority of the people in the Eastern countries possess an interdependent
concept of the self. Individuals perceive themselves in different ways which allows them to
identify themselves separately from others by taking in cues from the environment around them.
The norms and regulations encompasses the societal system and gets practiced by its members
through the various societal institutions and the societal groups they follow. These norms, ideals,
values and practices helps to shape the psychological understanding of themselves and the
environment around them. The development of the one’ self-construal is a process and is dictated
by knowledge based pathways called self-ways (Markus et al., 1997) which results in the formation
of the self.
10
The various self-ways can differ based on one’s cultural orientation as it has been developed
based on their cultural ideals, norms and acceptable behaviours affecting development of the self.
The cultural mandates that with time gets internalized and shape the psychological understanding
and evaluation of the world around form contrasting understanding of self-worth in different
cultures. Individualist cultures derive self-worth from being able to express themselves and
validate their internal attributes openly unlike members from collectivist cultures who seem to
derive self-worth from being able to maintain a harmonious relationship with their social group,
from being able to restrain themselves by being able to have the ability to adjust with the social
context. Individualist members pursue their own goals, happiness and personal fulfilment whereas
collectivist members believe in maintaining the expectation of others in their social groups,
maintaining relationship and group rapport (Markus & Kitayama, 1991).
Therefore, any change in the process of development of the self or any changes in the self-
construal process will have impact on our perception of the world around and will also affect one’s
behaviour. In a study conducted by (Gardner, Gabriel, & Lee, 1999), the causal role of culture they
examined the casual role of self-construal by priming independent or interdependent self-construal
within a culture and found that priming self-construal affected participants more when their prime
was inconsistent than when their prime was consistent with their self-construal having an impact
on their responses. The way we communicate, whether we use high context communication or low
context communication, for example, is also seen to be predicted by our self-construal and values
(Gudykunst et al., 1996). Our self-construal in turn helps to form not just our personal identity but
also our social identity. While such studies have been conducted a while ago, there is not enough
evidence to confirm that such changes have also been evidenced in the online domain. While
acknowledging the difference in behaviour due to cultural different selves (Hannover & Kuhnen,
2004; Choi & Choi, 2002) and values (Fiske et al., 1998) has been called for, through this research
it is argued that such cultural differences will also be evident in an online environment.
2.1.3. Social Identity Theory
Social Identity theory was developed to understand intergroup processes and relationships
(Tajfel et al., 1971). Through a series of minimal group experiments, Tajfel and his colleagues
11
found that members favour their in-group over out-group and there exists an intergroup
categorization when members of different groups come together. Competition between groups
further activates group membership further strengthening the bond with their in-groups, where
members strive to win against the out-group, which not only identifies their place within their
group but also helps to strengthen their group identity. While this research was conducted in an
offline context, it is argued that such intergroup behaviour is even more relevant in an online
context (Douglas & McGarty, 2001; Amichai-Hamburger, 2005) as social network extends from
the offline community to the online community (Ellison, Steinfield & Lampe, 2007). SNSs has
given its members the opportunity to become part of online groups, which fulfils their need to
belong to particular groups, which is a powerful motivation (Baumeister & Leary, 1995). The
motivation to become part of this wider community helps its members to form a common identity
with their group members, which generally leads to the formation of bond based identity over time
(Ren, Kraut & Kiesler, 2007). The absence of physical cues and the level of anonymity together
makes it even more important for members of groups to highlight their loyalty towards their group
as members strive to maintain their group identities ((Lea & Spears, 1991; see Klein, Spears, &
Reicher, 2007; Zhang, Jiang & Carroll, 2010).
Such behaviour is further explained in the social identity theory (SIT: Tajfel & Turner, 1979;
Tajfel, 1978), where it is argued that individuals interact in two ways: interpersonal and intergroup
communication. In interpersonal communication individuals are not aware of the social categories
and communicate in relation to the self. It is argued that such behaviour is relevant in an online
environment, where members communicate on SNSs and showcase their socially attractive
personal identity (Zhao, Grasmuck & Martin, 2008; Pempek, Yermolayeva & Calvert, 2009). In
contrast, during an intergroup communication there is a shift from the personal self to the group
self and members think and act in the benefit of the group, whereby members categorize
themselves in terms of the similarities with the in-group members and in terms of the differences
with the members of the out-group. Categorization also shift focus from thinking about the self-
concept in terms of one’s own personal values, attitudes and emotions but instead the self-image
of the members is then defined by their group membership. They further argue that by
demonstrating a competitive group behaviour make members feel secure and help them feel
positive about themselves. If this was to be seen in the context of a SNS environment, then it can
be argued that such group categorization certainly exists. Members interact with others with an
12
aim that they would be able to have a positive experience. Therefore, such existence of group bias
will exist.
Furthermore, maintaining a positive social identity is important for members especially when
they feel that there are in a group which has a low status quo as compared with the outer group
(Tajfel & Turner, 1979; Turner & Brown, 1978) by either leaving the group (physically or
psychologically), offer flattering to the in-group on positive group dimensions or even resort to
social change in order to help change the status of the group. Such behaviour is also relevant in an
online context as members engage in online communication in order to achieve a positive social
identity for themselves. While the main goal of being online is maintaining a positive identity for
the members, this could be argued to be more easily achievable in an online environment than an
offline environment (Barker, 2009). The level of anonymity on SNS allows its members to act in
ways that boosts group identity (Zhao, Grasmuck & Martin, 2008) however they are also free to
act in ways if they believe that the group norms or values are not in accordance with their own
norms or values by either becoming passive members (where they refrain from taking part in group
activities) or simply join other groups. For example, Reicher & Levine (1994) found that when
members identify with a powerful outgroup, participants seemed to lower those aspects of their
group identity which are inconsistent with the outgroup norms for the fear of punishment.
Additionally, SNS allows its members to have the advantage of being able to create a virtual,
editable social identity where they can highlight their positive attributes to others which might not
always be possible during offline communication.
2.1.4. Self-categorization Theory
The cognitive element of The Social Identity Theory (SIT: Tajfel & Turner, 1979; Tajfel, 1978)
which emphasise group interaction was further elaborated refined and moved beyond
understanding the intergroup processes to intragroup processes which was known as Social
Categorization Theory (SCT) (see Turner et al., 1987). Although SCT stemmed out from the same
assumptions as the SIT and while when we look at the social identity theory in whole, one would
look at both the SIT and SCT also acknowledging the differences in the emphasis of both theories.
While categorization was a fundamental component in the SIT, Turner and his colleagues extended
13
the concept of categorization and argued that the self-categorization theory not only worked at the
intergroup level but also worked at the intragroup level. They proposed that SCT was made up of
three levels that were important to one’s self-construal: the subordinate category of the self (or
human identity), the intermediate level of the self which is defined in terms of one’s group
membership (or social identity) and at a subordinate level one’s personal identity which is based
on intrapersonal comparisons with others and one level becomes salient the other levels becomes
less salient. One of the fundamental component of SCT is the process of depersonalization where
the member can relate and identifies with the social attributes of the group and in a way dictates
the attitudes, emotions and behaviours of the group members further resulting in group
polarization, i.e., when members of a group conform to the group norms (Mackie, 1986; Turner,
Wetherell, & Hogg, 1989) also found in an online environment (Spears, Lea, & Lee, 1990) which
is important for one’s personal identity (Zhang, Jiang & Carroll, 2010). However, personal
intention to join a particular group, and form one’s social identity is seen to influence online
interaction (Cheung & Lee, 2010) and can be argued to promote categorization. Unlike traditional
SIT, which suggests that group cohesiveness is a result of interpersonal attractiveness, SCT argues
that it is a result of depersonalization (Hogg & Hardie, 1991). Such level of depersonalization is
found to exist in an online environment where members aim to maintain a positive social identity
for themselves (Pempek, Yermolayeva & Calvert, 2009; Zhao, Grasmuck & Martin, 2008). SNSs
gives the opportunity to its members to find members with similar attributes, which is helped when
they become members of popular groups, helps them to enhance their social attractiveness to others
thereby contributing to enhancing their personal popularity. Members affiliate themselves to
popular group membership or act in ways which helps to boost their group membership that
benefit’s the individual’s psychological wellbeing, increase in social support, connectivity and
self-esteem (Ellison, Steinfield & Lampe, 2007) further supporting the concept of SCT.
SCT also help to explain group influence and power within a group. When a prototypical
member is able to influence others through his or her prototypical behaviour, there is an allocation
of power and status given to the member as they see the prototype as a leader through the self-
categorization at an intrapersonal level within a group (Hogg, 2001). Therefore, social
categorization is crucial in identification of the social self, which helps members to relate with the
group and thereby help to promote group influence (McNamara & Parsons, 2016) and group
behaviour (Turner, 2005).
14
While it had been argued that anonymity, group cohesiveness lead to deindividuation of the self
where members in a crowd loss self-awareness and behave in anti-social manner (Zimbardo,
1969). In contrast, (Reicher, Spears, & Postmes, 1995) argued that anonymity in a group instead
of resulting in the loss of self-awareness results in the identification of the self with the group
identity. They argue that the self consists of several levels (Turner et al., 1987) and when members
are in a group, their social identity becomes salient which helps them to relate to the group
membership. Interestingly, such online group members when used purposively and actively has
been seen to even effect offline civic and political views (Park, Kee & Valunzuela, 2009). The
model of SIT has been further applied in an online context and will be discussed in the next section.
2.1.5. Social Identity Model of Deindividuation (SIDE)
Unlike SIT (Reicher, Spears, & Postmes, 1995; Tajfel, 1978) and SCT (Turner et al., 1987;
Turner, Wetherell, & Hogg, 1989), which emphasizes the collective self, the Social Identity Model
of Deindividuation emphasizes the self. The SIT posits that members in a group perceive their
identity in terms of the identity with the group which is their social identity and group is
meaningful only when compared with other groups. Members attain a positive social identity
through positive group membership. Rewards will be more for in-group members (explained by
the minimal group paradigm) and in-group bias. This theory instead defines two selves, the
personal self which defines the person in terms of their personal attributes and social identity which
relates to the group they belong. In SIT, there is a shift from the personal self to the social self.
While SIT, emphasizes depersonalization where members in the group loose awareness of the self,
SIDE on the contrary emphasizes on depersonalization of the self, where the self is accentuated as
members are able to identify with the group. This concept of SIDE is closely related to SCT where
there is a process of depersonalization that helps members to identify themselves with others in
the group. It is argued that the online environment bear resembles to a crowd behaviour where
members act and behave in ways that is consistent to group membership. SIDE has been able to
extend the concept of SIT and SCT and extended this theory in computer-mediated communication
(CMC). Initially the concept of deindividuation was used to understand flaming in CMC and
researchers argued that anonymity in CMC lead to deindividuation and loss of self-awareness
which resulted in flaming and group arguments (Kiesler, Siegel, & McGuire, 1984; Siegel et al.,
15
1986). This has been contested in accordance with self-categorization theory, (Lea & Spears, 1991;
Spears, Lea, & Lee, 1990) and is argued that anonymity leads to group conformity as the social
identity of the in-group members are accentuated, thereby making them behave in line with group
norms. They also argue that when members of the group act in accordance with the group norms,
it helps the identity of the group to become salient when compared with other groups (Turner et
al., 1987). They further argue that when being anonymous or isolated, it can enhance group identity
and group conformity as they suggest that members of the group become more cognitively
attracted to the group as they feel that they can relate to group members more openly (Spears &
Lea, 1994). This is however in contrast to the traditional studies on group formation which is
dependent upon interaction with others.
In line with the SIT (Tajfel et al., 1971) which looked at the interaction between groups and
suggested that the groups were based on random criteria and did not have any history or personal
interest and intragroup interaction was not a salient feature but intergroup interaction was. This is
argued to remove any individual characteristics and because members are not aware of the other
in-group and out-group members, they treat everyone based on their group membership. This
behaviour can be seen in a CMC, where individual anonymity treat members based on their social/
group identity and not as individuals.
To summarize, the social identity model of deindividuation (SIDE) challenges social identity
theory (SIT) and social categorization theory (SIT) as it argues that deindividuation accentuates
social identity such that members in a group act based on their social identity and not their personal
identity which makes them conform to group norms. Building on the concept of SIDE, it is argued
that people make strategic use of these process (Klein, Spears, & Reicher, 2007) where the
visibility of the social self, influences how others perceive members such that it helps to affirm,
confirm and strengthen group memberships and also persuades group to adapt to specific group
behaviour. It is further argued that that such strategic behaviour influences group action which
could be found in intergroup as well as intragroup contexts. Unlike the concept of SCT, SIDE has
been able to suggest the importance of self-presentation in a group by complying with group
members, thereby maximizing the potential of being rewarded by group members (Baumeister,
1982; Leary & Kowalski, 1990). Similar to how members in a crowd lack individual identity but
instead identifies with the in-group leads to deindividuation, thereby helping members to establish
16
their social identity. If members on SNS are considered as members in a crowd, there too exists a
level of anonymity which leads to deindividuation, thereby helping members in the group on SNSs
to establish their identity in terms of group membership and also reinforce their intergroup identity.
Such behaviour is supportive of the SIDE, which is an extension of the traditional SIT and SCT,
is the theoretical base for this research.
In support of the past theories SIT, SCT and in particular SIDE, an attempt is made to use the
theoretical concepts in understanding online behaviour. While the use of SNS has been heavily
researched, there still exists a gap in the past research. Furthermore, while the concept of SIT and
SCT has been extended in the theory of SIDE, which helps to apply the theoretical concepts in an
online context, there exists a lack in the body of online research, which help to support these
concepts. While the online environment help to construct an online identity for its members, the
communication strategies advocated by SIDE still needs to be evaluated through empirical
research. In order to fully understand intergroup and intragroup behaviour researchers would need
to focus on using SNSs for conducting their empirical research if they want to understand online
behaviour. Using SNSs would help researchers to get access to intercultural population sample but
at the same time get access to real time data which is not always possible in other researches. It is
for this reason this current research will use Facebook as the communication tool which members
from all cultures uses (this will be discussed in Section 2.2. in detail in the below section).
Furthermore, as discussed in Section 2.1.1. and 2.1.2. acknowledging cultural differences is also
crucial if researchers want to fully understand online behaviour. The below section will discuss
the role of Facebook as an effective communication tool.
2.2. Literature Review (Psychological variables used in the study)
As Facebook is used as the communication tool to study differences in online behaviour, the
below sub-section will discuss the Intercultural importance of Social Media in general. This will
then follow with sections that will discuss the role of culture on the various psychological variables
that has been used as DVs in the thesis.
2.2.1. Intercultural Importance of Social Media
17
While the preceding paragraphs has been an attempt to provide the psychological perspective
on culture and the role of social identity in an offline context. This is important to highlight that
online behaviour can substantially be studied on social media, where Facebook being the most
prominent representative at present and is being used by members globally which has a diverse
population of its members (brandwatch, 2016; Statista, 2016). As of the second quarter of 2016
Facebook has over 1 billion active users worldwide and over 1.57 mobile active users (active users
represent members who had used Facebook in the last thirty days) (Statista, 2016). While 38.6
percent of the online population worldwide use Facebook, around 23.1 percent of online
population in Asia Pacific used Facebook at least once a month (Statista, 2016). Since its evolution
in 2004 Facebook has managed to create an environment which has allows members to interact
without any restrictions.
Understanding behaviour using online social media is unique in its own rights as such online
behaviour might not be seen or experienced offline. Although there are other ways of observing
interactive behaviours like focus group or experimental methods such methods are more
cumbersome. SNS like Facebook provides an environment which allows social researchers to
study human behaviour real time which had not been possible before, e.g., intragroup and inter
group behaviour, how members make friends and respond to others, along with many others. As
Facebook is being used across members from all cultures, studying human behaviour on Facebook
calls for ecological validity which might have not always been possible before. Additionally,
Facebook due to its immense popularity in its own right allows ground to investigate how members
on this online platform communicates, forms and maintains bonds. For many people use of SNS
has become part of their lives and it has led to the integration of our offline and online world which
has also seemed to provide psychological benefits (Ellison et al., 2007; Steinfield, Ellison, &
Lampe, 2008). One of the most common reason why people use Facebook is the desire to and
maintain relationship with their social capital (Ellison et al., 2007; Joinson, 2008; Lampe, Ellison,
& Steinfield, 2006; Sheldon, 2008). Facebook also allows members to distinguish between strong
and weak ties (Granovetter, 1973). It has also been seen that while members communicate with
their strong ties by posting comments or messages, they however seem to passively communicate
with their weak ties like just browsing around or just viewing the news updates (Burke et al., 2010).
18
The effects of the use of SNS is quite complex. While for some direct interaction on Facebook
seemed to reduce feelings of loneliness and increased feeling of social capital, others who
passively used Facebook where they just view contents of profile of their friends, like photos and
status updates seemed to report higher level of loneliness and reduced social capital (Burke et al.,
2010). Further research also suggests that there is a higher sense of pleasure for members who use
Facebook to extract information about others rather than just passively browse Facebook (Wise,
Alhabash, & Park, 2010). Such findings do seem to suggest that the use of SNS like Facebook has
complex psychological effects on its members which is moderated by different types of user
engagement. While Facebook has been used to eliminate boredom (Lampe et al., 2008) however
study conducted by (Pempek, Yermolayeva, & Calvert, 2009) suggest that irrespective of how
busy students were, they used Facebook for 30 minutes as part of their daily routine, suggesting
that boredom might not be the only factor for using Facebook. Motivating factors has also been
attributed to Facebook features, such as “birthday reminders”, which allow members to keep in
touch with their social sphere (Viswanath et al., 2009). In their study, they found that Facebook
features like birthday reminders helped 54 percent of users who interact infrequently were
motivated to use Facebook due to this feature further suggesting that motivation for using
Facebook is due to various reasons.
Impression management is crucial for its members as it allows others to perceive them online.
While various studies has attempted to understand whether the profile members present for
themselves is accurate or fabricated, have found that members fairly provide an accurate
information on their profile which reflected their personality (Back et al., 2010) suggesting that
members are not portraying an idealized self but an actual self on Facebook, which might be one
of the reason for its popularity as it has also been seen to be reflected of their offline characteristics
too (Weisbuch, Ivcevic, & Ambady, 2009). One of the reason why members portray a fairly
accurate image of themselves on their profile could be because as their contacts are mostly an
extension of their pre-existing offline relations (Ellison et al., 2006; Lampe et al., 2006), members
are aware as well as conscious that if they provide a fabricated self-presentation of their profile,
their offline members would identity it which will have an adverse effect on their social
relationship.
19
Cultural influence has also been seen to be affecting self-presentation on SNS. One study
showed that students in the USA would be more inclined to post inappropriate or problematic
pictures or information on their profile than German students (Karl, Peluchette, & Schlagel, 2010)
which supports the differences in cultural dimensions as suggested by (Hofstede, Hofstede, &
Minkov, 1991).
Identity preservation is also crucial in an online environment as it has been seen that members
often judge users based on the comments they receive from others (Walther et al., 2009) and
probably that might explain why users might judge members more attractive when comments are
left from attractive users rather than attractive users (Walther et al., 2008). This further explains
why members on SNS prefer to post attractive photos and comments about themselves as it has
been seen both males and females prefer to be friends with members who have attractive photos
(Wang et al., 2010). Maintaining an appropriate number of Facebook contacts is also relevant to
members’ profile identity as it has been seen that users are rated as attractive till about 300 friends
on their profile and then it declines (Tong et al., 2008) which might explain that just by having a
high number of friends might not be enough to influence other’s perception.
While SNS like Facebook is used by members from all cultures and social spheres, diversity in
social spheres has also been suggested to result in online tension (Binder, Howes, & Sutcliffe,
2009). On the contrary others (Lampinen, Tamminen, & Oulasvirta, 2009) has also suggested that
users seem to mitigate such online tension by utilizing the privacy controls which allows them to
limit access to certain friends, i.e., by communicating privately rather than publicly online or by
self-censorship, where they simply avoid posting pictures or information that might lead to online
tension. This in turn helps them to protect their social identity.
Use of Facebook has been seen to be having risks associated with it (Debatin et al., 2009) and
while members are aware of their privacy concern, they however seemed to trust Facebook and its
members more and hence agree to provide information when compared with other SNS like
MySpace (Dwyer, Hiltz, & Passerini, 2007; Fogel & Nehmad, 2009) which was also supported by
(Gross & Acquisti, 2005) where their study showed that in spite of being aware of the risks attached
members were still willing to disclose a large amount of personal data and only a small percentage
of members changed their privacy settings on Facebook.
20
Overall the information provided in the above paragraphs suggest important aspect of SNS as
a communication tool which has become part of our daily lives. While there is a wide variety of
research that has been conducted to understand the use of Facebook, there has been minimal studies
which has looked at the cross-cultural impact. This thesis hopes to address this gap in literature to
evaluate how culture can influence behaviour in an online environment. Some of the important
elements of communication on SNS involve members’ cultural orientation, i.e.., acknowledging
cultural diversity which has an effect on our psychological variables like the level of trust, level of
social support one receives concern for privacy, group identification and the level of self-
enhancement one gets while interacting. The below sections will review studies that has conducted
under each variable.
2.2.2. Online Group Identity
Members on SNS categorize themselves in terms of their group identity and perceives
themselves in relation to their group identity. This is in accordance with the social categorization
theory (Turner et al., 1987) which sheds its light on the influence of group membership. When
members identify with their group members, they no longer see their identity in terms of their
personal self but as a social self and their actions and behaviours are in line with their group norms.
This is certainly something which is quite often seen on SNS as members are loyal and supportive
to their in-group then out-groups. Such behaviour has also been seen to promote in-group bias
(Tajfel et al., 1971). As Facebook is an environment which is being used by members from diverse
cultural background, such bias behaviours could be speculated to be seen more among the
members from collectivist background then from members from individualist backgrounds which
can be explained by the cultural dimensions (Hofstede, Hofstede, & Minkov, 1991) when members
from collectivist countries as they try to maintain their group harmony which is one of the prime
goal of collectivists (Markus & Kitayama, 1991).
While the SIT explains intergroup process it however suggests that when members in a crowd
interact there is a loss of self-awareness and accountability (Tajfel & Turner, 1979; Tajfel, 1978).
However, (Turner et al., 1987) further extended this theory and explained that when members in a
group interact there is exists intergroup as well as intragroup process and this is certainly seen in
an online environment. However, as interaction on SNS lacks presence of physical, there exists a
21
certain level of anonymity in such environment. Interacting on SNS not only supported by the SIT
as members are able to identify their social identity online, but due to the level of anonymity, the
theory of social identity model of deindividuation effects (SIDE) comes to play, which is also an
extension of SIT and SCT (Lea & Spears, 1991; Reicher, Spears, & Postmes, 1995) and unlike
traditional SIT model, SIDE argues that anonymity accentuates social identity through
depersonalization unlike SIT which claims that anonymity leads to loss of self-awareness and
accountability. However, on the contrary SIDE is an extension of SCT which agrees that when
members in a group interact depersonalization of the self takes places which helps members to
identity with other in the group further leading to group behaviour. As members are unable to
identify their individual differences of in-group members, group identity accentuates. Although
members on SNS like Facebook are identifiable as they make themselves visible with the profile
information and pictures, the social identity of the group could be compromised. It is argued that
as members on SNS select their friends based on their preferences, such individual differences
would be fairly small. If, however when members would showcase their individual selves, it can
however lead to online tension (Binder, Howes, & Sutcliffe, 2009) as diversity in social spheres
could lead to online tension.
When members of other groups are identifiable, then it might also lead to social tension as
members can communicate directly to a particular member of the group and not the group as a
whole. If such direct communication is made at the group level, then it can also lead to intergroup
tension. For example, if a member on Facebook posts inappropriate information or gossip about a
member of another group, then as they are identifiable members, the members from the target
group can unite to respond back to the member. If, however the target group has a higher status
quo then it can dominate the communication and result in tension. Such intergroup unity might be
seen more among members from collectivist cultures as they believe in maintaining harmony with
the in-group to maintain their social status (Markus & Kitayama, 1991), however on the other hand
if members from the target group do not agree with the group action or behaviour then it can also
lead to members becoming passive. On the contrary members from individualist cultures unlike
members from collectivist countries might be free to move freely between groups. It can thus be
speculated that such collective behaviour might be more explicit among members from collectivist
cultures than individualist cultures. Alternatively, identifiability of outer group members can also
enhance social influence as it develops intergroup competition where members from both groups
22
strive to display their social status as it will undermine self-categorization with the group however
it will enhance intrapersonal differences within the group. If this is to be referred in terms of SNS
then when members interact with each other where there exists exchange of feedback from each
other, then such communication can also undermine group unity as it can lead to online tension
within group members. This is certainly in line with SIDE and SCT.
As there always exists a level of anonymity on SNS, members can develop their social identity
which is consistent with the group norms (Lea & Spears, 1991). While in their study members
showed behaviours that was consistent with their group norms, when they are anonymous rather
than when they are identified. (Spears, Lea, & Lee, 1990) also conducted a study which confirms
that anonymity lead to deindividuation which confirmed group polarization, further confirming
social influence in intragroup processes. Both these studies are in line with the SIDE theory and
also SCT. While these studies on anonymity support the cognitive consequences of SIDE, other
studies have also looked how members use different strategic processes to influence social
influence. For example, (Reicher & Levine, 1994) even found that when members identify with a
powerful outgroup, participants seemed to lower those aspects of their group identity which are
inconsistent with the outgroup norms and can lead to punishment. This finding is crucial when
trying to understand behaviour of individuals in an online environment. While it is important that
members maintain their social identity, they also need to understand their group status queue in
comparison to others as this has an influence on their behaviour. While members on SNS like
Facebook are identifiable, when members are exposed to other groups with powerful status then
based on previous findings, can undermine their in-group stereotype behaviour. For example, on
occasion when members might come across a group that might be more powerful when they fear
of punishment (e.g., work group or family) then members will generally undermine their usual in-
group behaviour and instead will behave in a manner that might be acceptable to the norms of the
outgroup. Probably this is why members on Facebook don’t usually disclose work related
information or refrain from discussing information that might have an adverse impact on their
offline relationship. Although it is speculated that such behaviour could also be influenced by
cultural orientation of the members, such that members from collectivist cultures would usually
portray such cautious behaviour higher than individualists when it comes to maintaining their
offline relationships like their family members. On the contrary when members are identifiable as
per SIDE it accentuates deindividuation which can result in members behaving in more
23
stereotypical manner (Douglas & McGarty, 2001). For example, as members on Facebook are
identifiable, they will be expected to act in accordance with their group norms as and use
communication strategies that are in accordance with their group norms. This sheds light to the
importance of self-presentation of the members of SNS which is demonstrated by contesting with
the out-group but it is also important to point out that with this identifiability, there comes a
heightened feeling of accountability to the in-group members. Identifiability of members can help
low status members to voice their opinions in regards to the outgroup as it will allow them to
become visible, thereby giving them the opportunity to raise their status within the group which
further helps them to self-enhance (Noel, Wann, & Branscombe, 1995). Such findings demonstrate
that SNS like Facebook acts as a platform which can also equip members with the opportunity
where they can display their identity amongst others. This can prove beneficial to members who
might not always get the opportunity to express themselves within their group or even during
offline communication, however due to the freedom one has during CMC, such members can
express and display their loyalty towards their in-group, thereby helping them to self-enhance.
Additionally, it can be argued that identifiability on SNS like Facebook can also have its drawbacks
as members might feel obliged to follow the norms of the group even though they might not always
agree to it and this can also lead decrease in self-enhancement. While such behaviour might be
explicit more amongst members from collectivists, it might not necessarily be something that
individualists might face (Markus & Kitayama, 1991) as due to their cultural differences, members
from individualist cultures might feel free to express their disagreement with the group norms if
they do not agree as they believe in being autonomous and independent. Additionally, such
expressive behavior on the other hand can also help them to accentuate their personal identity
which is in line with the cultural dimensions by (Hofstede, Hofstede, & Minkov, 1991).
Furthermore, as seen from past studies that identifiability also enhances feeling of accountability
(Douglas & McGarty, 2001), it is however argued that such accountability which can be a result
of identifiability will be more enhanced amongst members from collectivist cultures than members
from individualist cultures.
Therefore, overall it is argued that the increase in the use of CMC, i.e., SNSs for communication
purposes, intergroup and intragroup relations and anonymity in the medium play a crucial role in
understanding how members behave in such virtual environment due to the lack of physical
presence of its members which is one of the most crucial reason why it is even more important to
24
understand the role of identity in a group and how it has an impact on our behaviour. Below are
some of the hypothesis that this research aims to address:
H1: Intensity of Facebook use will mediate the relationship between culture and group
identity such that increase in Facebook activity will increase the level of group identity.
While Facebook allows its members to engage in different types of activities and the freedom
to use the medium whenever and wherever possible, such factors (type of activities and time spent)
will allow its members to identify other members based on their interests. Furthermore, spending
more time on Facebook will provide more opportunity to its members to identify with others.
H2: Group identification would be stronger among members from collectivist cultures than
members from individualist cultures.
Identification with others on SNS is a result of common interest and goals, which is most likely
to be achieved when members either know the other member from past offline relationship, or
have been able to share personal information between each other as it helps to identify with the
other member more than when they establish a new relationship online. The more information
members have about others in the group, the higher would be their identification with the group
members as they will be able to relate to identity of the group which helps in the development of
in-group trust among its members as has been supported by SCT. However, such trusting
relationship could be anticipated to be higher among members who might already have a shared
offline relationship, which could be speculated to be higher among members from collectivist
countries than members from individualist countries as they believe in maintaining their social
sphere more than members from individualist countries.
In summary, it is seen that group identity plays an important role on SNS. Due to the level of
anonymity that exists on SNS, identification with other group members helps members on SNS to
develop a social identity for themselves. While anonymity on SNS helps to develop one’s social
identity, categorization of members’ helps to maintain inter and intragroup relationship by using
different cognitive and strategic processes. The below section will now discuss online self-
enhancement.
25
2.2.3. Online Self-enhancement
People strive to hear positive factors about themselves and like to think about themselves as
better than others. This desire or motive to enhance one’s self worth is called self-enhancement
which is an evolutionary concept that allows personal and social advantages to people who possess
such traits (Sedikides, Skowronski, & Gaertner, 2004). This is not only relevant in our daily
interactions, i.e., when we interact with each other when offline but is an expression that members
from all cultures practice on SNS too. One of the main motive of using SNS is the advantage of
being able to have a sense of control over our self-image where members can persuade others to
perceive them how they want themselves to be perceived by others. Culture influences exists from
an early age (Wang, 2004) which makes people behave in cultural specific ways. For example,
people from individualist countries generally are high on individualism where members are
independent, open and free to express themselves and place themselves higher than the society and
values their personal achievement more and people from collectivist countries are generally high
on levels of collectivism where members believe in interpersonal relationships and are more
interdependent and place others higher than themselves (Hofstede, Hofstede, & Minkov, 1997;
Kitayama, Markus, & Kurokawa, 2000; Markus & Kitayama, 1991; Singelis, 1994; Triandis et al.,
1988).
The universality of self enhancement is however a debatable topic by many researchers. While
one body of researchers claim that self-enhancement is only found among Westerners and not
amongst people from East Asia (Heine et al., 1999). They argue that the need for positive self-
regard is not a universal concept but is more for people from Western countries whereas East
Asians like the Japanese are more self-critical. Such evaluations of East Asians have been
supported in various studies which suggests that East Asian are self-critical of themselves and thus
engage in self-criticism and not self enhancement (Heine & Lehman, 1999; Heine & Hamamura,
2007; Kitayama et al., 1997).
On the contrary other studies have argued that self-enhancement is a universal concept
(Sedikides, Gaertner, & Toguchi, 2003; Sedikides, Gaertner, & Vevea, 2005) where just like
Westerners, people from East Asia also self-enhance but their motivations are sensitive to cultural
context and are more tactical in nature, where Easterners self-enhance on collectivist attributes and
Westerners self-enhance on individualist attributes (Sedikides et al., 2005; Sedikides, Gaertner, &
26
Vevea, 2007). This was also supported in a study conducted by (Gaertner, Sedikides, & Chang,
2008) where they found that Taiwanese participants regarded themselves superior to their peers
on collectivist attributes than individualist attributes suggesting that self-enhancement is a
universal concept and is expressed more tactically among people in some cultures. However, it has
been further argued by (Kurman, 2001) that self enhancement is moderated by modesty and not
collectivist traits. In another study (Kurman, 2003) supports the universality of self enhancement
however it is argued that the differences in self enhancement is due to the differences in cultural
restrictions and modesty that is responsible for low self enhancement in collectivist cultures.
Members of collectivist and individualist cultures endorse values that are followed by their social
group and study suggests that high perceived cultural core values are more important than values
of high actual self-importance (Wan et al., 2007). Such theoretical concept if applied on SNS
would certainly help in understanding cultural differences on SNS as researchers appreciate the
cultural differences in the motivations of user engagement (Ross et al., 2005).
SNS gives people opportunity for instant feedback which allows members to self-enhance as
they actively seek positive feedback. Once members start receiving positive feedback their
enthusiasm to get more positive feedback increases than when they receive negative feedback
(Dunning, 1995). This could be suggested to strengthen the motive people have when they use
SNS as members not only increase their popularity but are also able to express a sense of
achievement. This also allows members to influence how they want others to perceive themselves.
However, it could be argued that such self-enhancement tactics where there is an open social
exchange amongst a diverse social group can sometimes lead to social tension (Binder, Howes, &
Sutcliffe, 2009).
In order to prevent such social tension on SNS members from different cultures behave in
different ways to preserve their relationships online. This also influences the extent to which
members create their profile and share information. SNS gives members the opportunity to create
an identity for themselves which gives them the advantage to influence the perception others
should be having for them which helps them to enhance their self-worth thereby contributing to
self enhancement. Members are able to create an online world which gives them the advantage to
have control on their self-worth. When members experience low mood, they can enhance their
self-worth by downward comparisons of others which helps them to manage their mood and helps
27
in self enhancement too (Johnson & Knobloch-Westerwick, 2014). Study conducted by
(Valkenburg, Peter, & Schouten, 2006; Zywica & Danowski, 2008) suggests that positive feedback
on SNS helped to enhance self-esteem and wellbeing of adolescent whereas negative feedback
negatively affected their self-esteem and wellbeing.
This need for positive self enhancement was also seen in different cultures as when asked to
rate themselves in comparisons to others people usually rate themselves higher than others on
positive traits (Brown, 1986) which again suggests that people from all cultures strive for self
enhancement however can depend on interpersonal relationships (Takata, 2003). He argues that
Japanese people who are from collectivist culture tend to be self-critical when they are under
competition free situation and when they share a personal relationship with others. However even
Japanese people displays self-enhancement traits, similar to North Americans (individualist
countries) when they come under a competitive situation and are with someone with whom they
don’t share any personal relationship. This is certainly an important point to address in this study
as people on a SNS come from both individualist and collectivist cultures and share the same
platform to communicate with each other. However while it has been argued that the online social
contacts are an extension of one’s offline social contacts (Acquisti & Gross, 2006; Ellison,
Steinfield, & Lampe, 2007; Lampe, Ellison, & Steinfield, 2006; Subrahmanyam & Greenfield,
2008) such ratings could vary as members might be influenced by their personal relationships they
might share with some of their group members. Even from an early age the need for positive
implicit self-enhancement gets instilled in us universally (Yamaguchi et al., 2007) and people self-
enhance based on norms and attributes that is supported by their culture (Brown & Kobayashi,
2002). This is certainly an important requirement in an online environment of SNS where
maintaining a harmonious relationship is important to maintain ties with the social contacts and at
the same time maintain one’s popularity among the group members.
Furthermore, it has also been argued that self-evaluation of the self is moderated by self-esteem
(Brown & Mankowski, 1993). In their study, there were two groups who had to rate themselves
and both group of respondents viewed themselves as more favourably when they were in a positive
mood and members with low self-esteem members lowered their self-evaluation scores when they
were in a negative mood. It was further suggested that such self-depreciation tendency where
people perceive themselves negatively can lead to psychological impact like distress (Kurman,
28
2003). Kurman further suggests that self-enhancement helps to develop self-esteem which also
helps in the wellbeing of individuals in both cultures. In a study by (Zywica & Danowski, 2008)
it was also found that people with high self-esteem supported their self-enhancement hypothesis
where they were popular among both their offline and Facebook group. Interestingly people who
were more introvert and were less popular offline, showed low self-esteem and supported their
social compensatory hypothesis where they worked harder to become more popular on Facebook.
This is a very important point to be addressed as members on SNS who might be popular offline
will also strive to maintain their popularity online and people who might not be so popular among
their social group offline will also strive to become popular online suggesting the advantage to
communicate with others without having any social pressures. It is argued that such benefits of
SNS might be since it lacks face-to-face cues and physical presence which gives advantage to
people who might be introverts. It gives them the advantage to communicate in a one to one way
with members they are more comfortable with. It also gives them the advantage to give and receive
social support whenever required thereby contributing to their positive self enhancement and
wellbeing.
While one group of researchers claim that members who have established offline contacts self-
enhances by extending their social contacts online too and has been hypothesized as Social
enhancement (“Rich get Richer”) or the “Matthew Effect” (Merton, 1968). Whereas the other
group of researchers suggest that it is a Social compensation method (“Poor get Richer”) where
when members have failed to establish adequate offline social contacts compensates by
establishing online social contacts (Barker, 2009; Kuss & Griffiths, 2011). This is in line with the
SIT as it is seen that when members fail to identify with their social groups they look for new
social relationships. It is also suggested that there will be an increase in self-categorization when
members fail to identify with their established social network. Such identification and
categorization is speculated to be more on SNS due to the level of anonymity members have.
Members are thus free to create their own social groups and even maintain privacy settings which
allows members to interact with other members without any social pressure. While this could be
seen to have its benefits, however if members are not careful this can often lead to online social
tension if the members they might be newly interacting with, might also be linked to the other
members they might not want to connect or interact with. Although there might be risks attached
29
to such process, the benefits that come with it can certainly be linked to help in online self-
enhancement.
The overall theoretical literature provides a number of important insights on the importance of
self-enhancement in an online environment like SNS. The universality of self-enhancement tactics
used by members from both individualist and collectivist cultures is supported in this study. Past
studies as seen has mostly focused in understanding the role of self-enhancement mainly in an
offline context or looked at self enhancement in relation to personality traits or demographic
differences. There is a lack of studies that has looked at the role of cross-cultural impact on the
role of self-enhancement in an online environment and this research fills in this gap by evaluating
the cultural influence on self-enhancement in an online environment. Below are some of the
hypothesis which this study aims to address:
In the following study, it is hypothesized that:
H3: Forms of online self-enhancement will vary across cultures. Individualists self-enhance
more on individualist attributes compared to others online whereas collectivists self-enhance
more on collectivist attributes compared to others.
Self-comparison has been seen to increase sense of competition even in collectivist countries.
Therefore, it is speculated that even though members from collectivist cultures will have more
social contacts that they might share an offline relationship with, when members are compared
with others they would also prefer to place themselves at a higher place. However, it is expected
that members would rate themselves higher on their cultural specific traits even when they are on
SNS.
H4: Intensity of Facebook use will mediate the relationship between cultural orientation of
participants and their online self-enhancement such that more intense Facebook use leads to
more self-enhancement in both cultures.
The type of activities members undertakes and the time they invest helps members to increase
the chances to interact with their social contacts on Facebook. However not all members will
interact in the same manner and the level of participation and interaction with the social group will
vary based on their goals, motivations and also cultural constraints. It is envisaged that in order for
members to self-enhance they need to interact with their social contacts where differences in
30
activities and frequency of Facebook use will have an impact on the level of self-enhancement and
members from collectivist countries (Indonesia) will self-enhance more on collectivist attributes
than individualist attributes. Members from individualist countries (UK) will self-enhance more
on individualist attributes than on collectivist attributes.
H5: There will be a positive relationship between network size in both cultures and online
self enhancement such that increase in number of social contacts on Facebook will increase
feelings of self-enhancement for participants in both UK and Indonesia.
While it is important to understand that one cannot communicate in isolation, therefore it is
envisaged that participants will be able to self-enhance more with the increase of their social
contacts. Although level of self enhancement will vary with cultures, it is speculated that members
from individualist country (UK) will self-enhance more than their counterparts as it is expected
that they will have a higher number of social contacts on Facebook.
2.2.4. Online Perceived Social Support
SNS provides an environment for its members which helps them to interact with their social
network without any restrictions. Past studies have found improved life satisfaction associated
with Facebook use (Ellison, Steinfield, & Lampe, 2007; Valenzuela, Park, & Kee, 2009). Being
able to receive instant feedback is something that was not possible before and when members are
able to receive positive, it has resulted in the increase in self-esteem and well-being. Alternatively,
negative feedback from members on SNS has also seemed to decrease self-esteem and wellbeing
(Valkenburg, Peter, & Schouten, 2006). While SNS like Facebook helps members to extend their
social relationships from offline to online (Ellison et al., 2007), it also provides its members
increase opportunity to seek and provide social support. Additionally, SNS also allows its members
to meet people with similar interests and goals (Ellison, Heino, & Gibbs, 2006) where members
have the advantage to extend such online relationship to offline relationships (Parks & Floyd,
1996), suggesting that members are not just using the SNS for communication purposes but also
extending their offline social spheres.
Being able to meet new people online, members can compensate for the social support people
received offline as they are able to meet people online with whom they could identify with
31
(McKenna & Bargh, 2000). Human beings are social animals and the importance of our bonds
with friends cannot be ignored as it contributes to our happiness (Myers, 2000). SNS like Facebook
provides platform for its members, where they can maintain existing relationships but also create
new ones (Ellison et al., 2007), they can also create an online profile of themselves with
information that they believe would persuade others to perceive them how they want to be
perceived, thereby helping members to boost their self-esteem (Gonzales & Hancock, 2011).
While SNS, allows its members to expand their social network which might have not been possible
before which allows them to seek social support whenever required from a wider network,
however, it could also be argued that larger and diverse social sphere might consist of close ties,
who might provide social support when in need. While the number of friends might be higher for
individualists who are not restricted by their commitment to their social group but instead has the
advantage to make decisions freely, the number of friends on SNS for members from collectivist
countries might on the other hand might be mostly restricted to close network of friends which can
have an impact on the support one receives on online.
Social support has been seen to be positively associated to the number of friends and subjective
wellbeing (Myers, 2000). Having too few friends can limit the chance of social support one can
receive. While SNS like Facebook provides the platform for its members to create their own social
sphere where members can friend other members based on their interests where members can
increase their social capital. However, the social support members receive would depend on the
level of trust members have between each other. When members have larger social network, it is
not always possible to invest their time with all its members in the network. Therefore, they might
not have the advantage of receiving social support from all its members on the social sphere.
Members might have simply added new members to their social sphere to display their popularity
and not just for relationship building. On the contrary, having too many friends on SNS has also
been seen to have a negative impact on its member’s popularity (Tong et al., 2008) suggesting that
if members show a larger than real number of friends on their network it can instead draw them
away from each other. Therefore, it is important to understand that if members want to maintain
online social networks from which one can seek social support in times of need or crisis, it is
important to maintain a realistic number of friends on the social network which can vary between
cultures.
32
Members on SNS like Facebook are able to communicate with each other via different means,
such as private messaging, commenting and even via their profiles (Ellison et al., 2007). The
advantage of being able to add editable information on Facebook, allows its members to display
information selected individuals or selected group can see (Walther, 2007), which help members
to portray their positive attributes of their identity. It is therefore argued that this editable
functionality on SNS like Facebook help accentuate attributes of their social self which is
important in maintain a social identity online (Lea & Spears, 1991; Spears & Lea, 1994) as
members shift their thoughts and behaviours by focusing on social self rather than their personal
self. This is a advantage for members on SNS as it allows members to show that they could be
trusted which is displayed by their profile information online, that help to form their online identity
(Tajfel et al., 1971). Furthermore, by being able to self-categorize themselves to their respective
groups members are able to portray their group identity to others (Tajfel & Turner, 1979; Tajfel,
1978; Turner, Wetherell, & Hogg, 1989) which could be argued to provide members the advantage
to display group oriented behaviour (Klein et al., 2007), which help members to show their loyalty
towards their group but also increase their opportunity to seek social support whenever required
which however could vary due to cultural differences. While it could be argued that such
functionality would be beneficial for members who join SNS to seek support which they might not
get offline, it is however important to portray truthful representation of themselves, if members
want to maintain long-lasting and trusting relationship. (Gibbs, Ellison, & Heino, 2006) where
members not only get subjective well-being of themselves but also helps members to receive more
social support (Ko & Kuo, 2009; Zhao, Grasmuck, & Martin, 2008) and as such can also equip
members to deal with negativities in life (Dunn et al., 1988).
It is however important to understand the effect culture can have on the self-presentation of
members on Facebook. Having an existing offline relationship with online members’ result in the
display of truthful representation of themselves, as members might fear that their offline contacts
would become aware of any false representation, which can have adverse impact on their social
identity. As this is an anonymous environment, where members do have the privilege to indulge
in one-to-one private communication with other members, the fear of having a bad reputation
among the social network would refrain members from posting false or inaccurate information on
SNS. However, this could certainly be achieved if members create pseudo profiles for themselves
33
or create an alternative profile which with selected members who they want to display exaggerated
or inaccurate information.
This section outlines the importance of social support in an online environment. While SNS is
helping members to access and extend their social network, it has also been seen to have its benefits
(Ellison et al., 2007). It has also been highlighted that by providing a truthful visual information,
members can manage their identity which in turn give members them the opportunity to seek social
support, when needed. This is particularly beneficial for members who might find it difficult to
express themselves when communicating face-to-face. Anonymity on SNS is therefore helpful in
providing a platform which help members to freely express themselves as it lacks the physical
presence of the members. Therefore, it can be argued while face-to-face communication might be
better for some as it helps to seek social support directly, anonymity on SNS on the other hand
helps to provide a platform where members can use their discretion freely to seek social support
from members they are comfortable with. Therefore, understanding the role of social support in
an online environment is crucially important if researchers want to understand online behaviour.
In order to understand how members in different cultures views perceived social support online,
below hypotheses were outlined:
H6: Intensity of Facebook use would mediate the relationship between culture and social
support, such that there increase in Facebook intensity would increase perceived social
support for its members in both cultures.
Members from collectivist countries believes in maintaining a harmonious relationship with
their social sphere which they consider as the prime goal for themselves unlike individualists who
believes in being independent and autonomous (Markus & Kitayama, 1991f; Markus & Kitayama,
2010) would be seen to have an impact on the time and frequency of Facebook as due to social
restriction and availability. This would be having an impact on social support members receive
online however on the contrary as members from collectivist countries would have higher number
of friends on their social sphere with whom they might already have established relationships
offline would result in members from collectivist countries would receive more social support than
members from individualist countries.
H7: Increase in Facebook network size would have a positive impact on perceived social
support that members receive online.
34
As discussed in previous paragraphs, network size seems to play an important role on social
support. While increase in the number of friends helps to increase the possibility of being able to
get access to a wider network from whom members have a higher potential of seeking social
support, however it is also important that the network size is realist and not inflated which has
been seen to also have negative effect however cultural differences would have an effect.
2.2.5. Online Trust
Individuals prefer to disclose more information when face-to-face than when they interact
online (Emanuel et al., 2014). Anonymity on SNS, which lack presence of physical cues calls for
a level of trust between members which develops between members with time (Wilson, Straus, &
McEvily, 2006) as they engage in activities online, thereby helping develop a trusting relationship
between each other that motivates online communication (Fogel & Nehmad, 2009; Valenzuela,
Park, & Kee, 2009). This development of trust, without having any physical cues is something that
has fascinated social scientists. While on SNS members can create their online profiles for
themselves with personal information about themselves such as name, date of birth, contact details,
personal interests and add their pictures that are editable. It is possibly these information cues
about other members that persuades others online to trust other members. Members on SNS
indulge in online communication despite being aware of the risks associated with it. It can be
speculated that as members in an online environment lack the face-to-face connection, they are
forced to implicitly trust their social contacts on SNS like Facebook and assume that other
members will not abuse their personal information. This virtual relationship which develops due
to their shared beliefs and interests among each other helps them to develop such trusting beliefs
(Walczuch & Lundgren, 2004) which helps to mitigate such risks in an online environment. Being
able to see information about others, it helps them to develop a feeling of trust which makes them
disclose information about themselves too. The world of SNS is an environment which members
cannot control as members are free to share their opinions and views without any restrictions.
However due to the level of functionality which equips members to have control of their
information helps in the development of trust as it helps to mitigate such perception of risks
associated with such online communication. For members to trust each other, they would need to
persuade others to trust them which is in line with the identity performance function of SIDE,
35
fulfilling the identity consolidating function (Klein et al., 2007) which further helps members to
persuade others to follow them.
By providing personal information on the profiles, members can reduce ambiguity on SNS and
help to create positive interpersonal relationship with other members (Tanis & Postmes, 2003).
When members are not aware who they are interacting with on SNS members develop a sense of
uncertainty as they are unable to predict the behaviour of the other person which results in less
trusting perception for the members. It is however anticipated that as members from collectivist
countries might have a higher number of social contacts on their Facebook network with whom
they might already have established social relationships, the feeling of uncertainty might be less
among collectivists members than members from individualist countries as their social network
might be more diverse and they might have less number of members that they have established
offline relationships.
Members on SNS friend others and engage in communication only when they are able to
establish common interest among each other which help them to develop an intrapersonal and
interpersonal relationship online which is in accordance with SIDE (Lea, Spears, & de Groot,
2001; Postmes et al., 2001; Sassenberg & Postmes, 2002) which postulates that unlike the de-
individuation model members in a group are able to identify with the other members and
accentuate group membership. This identification on SNS with their in-groups and even out-
groups helps to develop a level of trust and reciprocity. While it has been argued that the level of
trustworthiness and reciprocity will be higher for in-groups thereby giving rise to in-group
favouritism and prejudice behaviour among group members. In respect to SNS, identifiability with
other members is therefore crucial which helps to develop a trustworthiness towards others,
however in accordance to the SIDE Model such level of trustworthiness is accentuated towards
group membership when identity of members is not available. However, on SNS like Facebook,
identities of members and group members are available as members have their profile information,
it is therefore expected that such level of trustworthiness will be higher towards group membership
as this will help to show group loyalty in view of reciprocity. In accordance with the SIDE Model
members on SNS interacts with an expectation of reciprocity from other group members however
this can vary based on cultural orientation of members. While members from collectivist cultures
might be more inclined to show trusting behaviour and reciprocity towards others as they might
36
be having established offline social contacts with the other contacts. This will help them to show
their level of loyalty towards the other member with a hope that such behaviour would be returned
for them in the future too. However, such reciprocity behaviour towards individual group members
would only be possible when it has positive impact for the group. On the contrary members from
individualist countries might also show reciprocity towards individual group members however
only when it might accentuate their own personal identity and help to stand out from other in-
group members further highlighting the role of culture on user behaviour.
Due to the level of anonymity, members strive to form a bond by providing personal
information about themselves that helps to reduce uncertainty (Hogg & Turner, 1987), which helps
to build trust among each other (Metzger, 2004) and results in exchange of communication, which
highlights the requirement of mutual exchange of information further supporting the Social
Exchange Theory (Roloff, 1981) which posits that if social exchange takes place when there is
perceived benefit in any particular interaction and is measured in terms of cost. Trust plays a
central role in Social Exchange theory as high level of trust leads to low perception of cost and
vice versa. Therefore, within a SNS environment members would seek out relationships with other
members by looking at the personal profile information and would involve in positive interaction
and would refrain from interacting with members that would have negative outcome for them.
Once members on SNS can identify with their group they are able to develop a trusting relationship
with their group members. Additionally, when there is any threat to the identity of the group, it
also leads to in-group bias or prejudiced behaviour towards the in-group members (Turner et al.,
1987; Turner & Oakes, 1989; Voci, 2006). Due to the cultural differences, it is anticipated that
members from collectivist cultures, who would have a stronger group identity (as their groups
would generally consists of members with whom they might already have offline relationship)
would show prejudiced behaviour towards their in-group members more than members from
individualist cultures which helps members to show loyalty towards their group which helps
members to achieve self enhancement (Sedikides, Gaertner, & Toguchi, 2003; Sedikides,
Gaertner, & Vevea, 2005). It is further anticipated that such trusting behaviour would be stronger
among members from collectivist cultures than members from individualist cultures. On the
contrary, it has been argued that by having a larger social capital gives members the opportunity
to trust more people and develop relationships further (Realo & Allik, 2009) as in their study they
37
have found that individualists tend to have a larger social group unlike collectivists as they like
maintaining their trust among their own-group.
For members to know information of their in-groups or even out-groups, members are required
to invest their time and often exchange communication with each other, which requires them to
invest time and make an effort to interact with others. While such involvement might have positive
outcomes like increase trust (Valenzuela et al., 2009) but has also been seen to decrease student
cognitive performance (Rouis, Limayem, & Salehi-Sangari, 2011) and excessive use of the internet
has also been linked to addiction (Kuss & Griffiths, 2011). Alternately it has been also argued that
the type of communicative activity is based on the individual’s motive or goal of whether one
wants to self-promote or want to maintain relationship (Underwood, Kerlin, & Farrington-Flint,
2011). Although their study was conducted on undergraduate students, this could be well extended
to see how cultural self-construal of members would have an impact. However further studies
confirm that that members from collectivist cultures would be more inclined to maintain old ties
and members from individualist cultures would be more inclined to self-promote based on their
cultural difference (Ji et al., 2010).
This section highlighted the importance of trust in develop one’s social identity. While trusting
relationships are easier to form in an offline world due to the physical presence as members are
able to predict the behaviour of others, this is even more important in an online environment where
there exists a level of anonymity. Being able to develop a trusting inter group and intra group
relationship helps to maintain an online social network. This is seen to been achieved by members
disclosing personal information about themselves by wall post, pictures, status updates on their
profiles which helps to reduce uncertainty allowing members to successfully predict their future
behaviour. Such behaviour in in line with SIDE and the Social Exchange Theory. However, as the
world of SNS consists of members from around the globe understanding cultural influence on the
level of trust and user behaviour is called for which will be analysed through the below hypotheses:
H8. Intensity of Facebook use will mediate the relationship between culture and online trust,
such that more intense Facebook use higher use will increase online trust in both cultures.
For members to develop a trusting relationship it is important that members disclose personal
information which helps to reduce uncertainly and allows their social contacts to predict their
behaviour. Such trusting relationship is also attained through various activities on Facebook and
38
the time members spend online. It is anticipated that the higher members use Facebook the more
opportunity they get to develop a trusting relationship with their in-group members and vice versa.
H9: Levels of trust would differ due to cultural differences such that collectivists would be
more trusting towards collectivist members than individualists.
While members from collectivist countries prefer to maintain their in-group identity by restricting
their social contacts to generally known members, there is a higher expectancy that as they are
known members, whom they mostly share an offline relationship too, it thus becomes easier for
members from collectivist countries to predict the behaviour of its in-group members than
members from individualist countries as their social network online could have a higher number
of friends whom they have accepted online, that they might know offline due to which it might be
difficult for such members to predict their behaviour. Therefore, it is expected that the level of
trust will be higher would be among members from collectivist countries than individualist
countries.
2.2.6. Online Privacy Concern
The online environment on SNS like Facebook require its members to share information with
others that will allow others communicate with each other facilitating new connections (Lampe,
Ellison, & Steinfield, 2006). Sharing personal information has become one of the required norm
to help increase social capital. Members not only disclose their names and occupation but they
also disclose their date of birth, hobbies, interests, sexual preferences, tastes in movies, music and
political views (Gross & Acquisti, 2005). It is now a cultural norm to upload photos and selfies on
user profiles to seek instant attention from social groups. While members are aware of the potential
online risks, it has however been argued that the benefits achieved by sharing such personal
information outweighs the potential privacy risks (Culnan & Armstrong, 1999; Dinev & Hart,
2006), such as the enjoyment one achieves (Rosen & Sherman, 2006), self-presentation (Boyd,
2007) and maintenance of social ties (Ellison, Steinfield, & Lampe, 2007) to receiving social
support (Bender, Jimenez-Marroquin, & Jadad, 2011) are some of the benefits of the relationships
shared among their social group which outweighs the potential privacy risks equipping members
to judge the benefits and costs of the interaction which is supported by the Social Exchange Theory
(Homans, 1958). Privacy concern and potential risks on SNS has been a datable topic for a while.
39
While on some occasions members might be interested in showcasing information to only their
own social group whereas sometimes, they might want to reveal information to anonymous
strangers and not to known people. Information revelation offline is certainly different from
revelation of information online. The subjective relationship that we share with our offline
contacts, we categorize our social contacts as “strong or weak ties”, such categorization is not
always so straightforward on SNS as members on SNS can only categorize their contacts as either
“friend or not”. Their acceptance as a friend can include members in the group that they might be
comfortable with. This can certainly raise concerns at a personal level as members might not
always be comfortable to share personal information to all members of the group and this is when
members either become cautious of what information they post online or become passive group
members.
The social spheres members share, are an extension of their offline contacts helping them
maintain relationships (Ellison et al., 2007). SNS, like Facebook allow members to spy on their
members, where they collect information about others they are interested in, allowing them to
know more information about their social contacts rather than just add new social connections
(Lampe et al., 2006; Pempek, Yermolayeva, & Calvert, 2009). It has further been argued such
surveillance which allows members to know about their social sphere helps to increase Facebook
activity. Possibly that is why members prefer to use less privacy settings (Gross & Acquisti, 2005).
Their study suggests that 1.2% of users changed the “default” search privacy setting and less than
½% changed their “default profile visibility” option. One of the possible reasons why members
feel that in spite of the risks to their privacy they are unconcerned about the use of privacy settings
due to the fact that it restricts their opportunity for others to view their information. It is argued
that such behaviour is in accordance with the Social Exchange Theory (Homans, 1958) as members
feel that everyone online is in the same situation and by not restricting their information would
help in reciprocity by others. This will further help them to get information about each other,
thereby helping to build trust (Metzger, 2004). While members are ready to take risks online on
how much information they want to disclose, such information disclosure also depends on the
context of online spaces as it has been found that members would disclose more information in a
generic context than when asked to share information on particular context (Emanuel et al., 2014).
This highlights that there exists a level of cautiousness among members using the online platform
40
which is possibly why they also found that members prefer disclosing more information face-to-
face than online.
Members on SNS tend to sacrifice their concern for privacy for the benefits that come with it
by weighing the costs and benefits that come with it (Hui et al., 2006). Additionally, information
fairness helps to build trust that can also reduce privacy concern (Culnan & Armstrong, 1999).
Disclosing information on SNS is a crucial requirement for exchange of communication,
development and maintenance of any relationship as it help members feel that the other individual
is not a stranger. As humans, we tend to become anxious when we are unaware of the situation or
when we are not able to predict the behaviour of others due to the lack face-to-face interactions
and the physical presence of others, which is why there is always a level of uncertainty in
perceiving how the other person might be feeling or thinking. However, when members disclose
information about themselves it helps to relate to other members, which help in the reduction of
uncertainty, as members can predict their feelings and actions.
Being able to get information about group members help in the identification with other group
members, which leads to depersonalization and assimilation to group norms (Turner et al., Hogg,
1987). However, depersonalization also lead to in-group bias. The use of filter settings on SNS,
like Facebook can also be argued to be a result of depersonalization, which is supported by SCT
model, which posit that, when members become part of certain group there is a shift from personal
self to a group self and their behaviour is directed towards group goal. Based on such findings it
can be argued that such categorization of self would be more explicit among members from
collectivist backgrounds as they might be more cautious towards privacy risks than members from
individualist countries and thus will be more inclined to use filter settings on Facebook.
Furthermore, as members can control what information to disclose and who can view the
information, has also been said to reduce privacy concern (Krasnova et al., 2010). Trust among
internet providers has also been seen to reduce such privacy concern, as members can have control
of their information they want to disclose. Interestingly their results failed to show that trust among
members on SNS helped to reduce privacy concern. In addition to trust internet interest has also
been said to reduce perception of privacy concern (Dinev & Hart, 2006). While trust seem to
moderate privacy concern, the role of culture certainly cannot be ignored as in a study conducted
by (Akbulut & Günüç, 2012) it was seen that using nick names and not full names lead to less
41
social support from friends and members also received less social support when they added
unknown members on their social sphere online further confirming the importance of personal
information and online privacy. Their study was conducted in Turkey, which represents members
from collectivist countries, their study successfully sheds light on the importance of online privacy
among collectivist cultures.
Looking at it from a cultural perspective we would however argue that members from
collectivist cultures would be more cautious about disclosing their personal information for the
fear of social tension among their social group which might also have an adverse effect on their
offline relationship. However, past study conducted by (Binder, Howes, & Sutcliffe, 2009) also
suggests that diversity of social spheres predicted online tension. In their study on Facebook
participants reported that common features on Facebook which allows members easy broadcast of
information lead to online tension. They further suggest that online tension is more likely to be
between social spheres and less likely within spheres, suggesting that when offline relationships
are extended online there are higher chances of online tension. This is certainly crucial for
members who has a higher number of offline contacts on SNS, such as members from collectivist
countries, as there would be higher chance of online tension between the social spheres among
collectivist members than among members from individualist cultures, which might shed light on
the reason why members from collectivist cultures are more cautious in their communication and
maintain a lower network size. It is therefore argued that the maintenance of privacy settings would
be more crucial for such members if they want to avoid online tension as it is not always possible
to have control of what others might disclose.
Therefore, based on past research and theories it is known that concern for privacy does
influence how members behave online. However, by being able to develop a trusting relationship
where members are given fair information by network providers that facilitate information control
to the members on SNS can help eliminate such privacy concern. Members are able to weigh the
potential costs and benefits from the interaction and make informed decisions. As SNS is a
multicultural platform of SNS which have differences in how members perceive such privacy
concern that needs to be evaluated if researchers want to understand online behaviour. The below
hypotheses are hoped to answer some of the questions in the study:
42
H10: There would be a negative relationship between intensity of Facebook use and online
privacy concern, such that increase in Facebook activities and Facebook frequency would
decrease privacy concern among online users.
While it is important for members to interact with others to develop and maintain their online
relationships, it is envisaged that the increase in FB activities (type) would result in decrease in
privacy concern among SNS users as they would be able to know personal information about other
members they interact with, which helps in uncertainty reduction.
H11: Privacy concern will be more explicit among members from collectivist cultures than
members from individualist cultures, such that collectivists would have more privacy
concern than individualists.
While disclosing personal information is crucial in helping members reduce feeling of uncertainty
which helps them to maintain a trusting relationship supporting Social Exchange Theory (Homans,
1958) where members can also accept such trusting behaviour from others too.
2.3. Summary
This section was an attempt to look back at past research outlining the theoretical basis of this
research which looked at SIT, SCT and SIDE but also outlined the importance of SNS. Cultural
orientation bears particular importance in various psychological constructs like online group
identity (which helps members to develop their social identity), self-enhancement (helps members
to self-enhance due to the interaction they make with their social network online), perceived online
social support (is the support that members receive when in need), online trust (helps members to
develop and maintain trusting relationships) and online privacy concern (allows members to
maintain their privacy). It is believed that to fully understand online behaviour through the eyes of
culture, it was important to evaluate each of these constructs in one single research, which has not
been attempted before. It is hoped that the hypotheses under each psychological construct would
allow readers to understand the causal role of culture in that specific psychological construct.
While culture is used as the independent variable (IV), online group identity, online self-
enhancement, perceived online social support, online trust and online privacy concern are used as
dependent variables (DV) in the study. As Facebook provides the advantage of getting access to
43
an inter-cultural population, this has been used as the platform to access the constructs named
above. While the use of cross-cultural research using native population, sample is scarce, this
research would not only use all the psychological variables in one study but will also use native
sample population to evaluate the true cultural differences in the responses, further helping to
maintain validity and reliability of the data. Therefore, this study was conducted in India and
Indonesia (where members generally higher on collectivist attributes) and in UK (where members
are generally higher on individualist attributes). The below diagram Figure 2.1. illustrates the
theoretical framework designed to evaluate the aims and hypotheses in this study. The figure
illustrates how the outcome variables like social identity, self-enhancement, social support, trust
and privacy are effect by the cultural differences of its members based on how they engage on
social media, like Facebook.
Figure 2.1.
Visual Representation of the Main Constructs used in the Study
44
2.4. Research Aims and Hypotheses
This section outlines all the research aim and hypothesis for all the constructs in order to facilitate
reader understanding.
Aim:
a) This research evaluates how differences in cultural self-construal affect online
communication, the decisions SNS members make online and the psychological outcomes of
these processes.
b) Additional hypotheses were used to evaluate how Facebook management has an impact on
psychological outcome variables, independent of culture.
Below are the hypotheses which the project aims to address:
Online Group Identity
H1: Intensity of Facebook use will mediate the relationship between culture and group identity
such that increase in Facebook activity will increase the level of group identity.
H2: Group identification would be stronger among members from collectivist cultures than
members from individualist cultures.
Online self-enhancement
H3: Forms of online self-enhancement will vary across cultures. Individualists self-enhance more
on individualist attributes compared to others online whereas collectivists self-enhance more on
collectivist attributes compared to others.
H4: Intensity of Facebook use will mediate the relationship between cultural orientation of
participants and their online self-enhancement such that more intense Facebook use leads to more
self-enhancement in both cultures.
H5: There will be a positive relationship between network size in both cultures and online self
enhancement such that increase in number of social contacts on Facebook will increase feelings of
self-enhancement for participants in both UK and Indonesia.
45
Perceived online social support
H6: Intensity of Facebook use would mediate the relationship between culture and social support,
such that there increase in Facebook intensity would increase perceived social support for its
members in both cultures.
H7: Increase in Facebook network size would have a positive impact on perceived social support
that members receive online.
Online Trust
H8. Intensity of Facebook use will mediate the relationship between culture and online trust, such
that more intense Facebook use higher use will increase online trust in both cultures.
H9: Levels of trust would differ due to cultural differences such that collectivists would be more
trusting towards collectivist members than individualists.
Online Privacy Concern
H10: There would be a negative relationship between intensity of Facebook use and online privacy
concern, such that increase in Facebook activities and Facebook frequency would decrease privacy
concern among online users.
H11: Privacy concern will be more explicit among members from collectivist cultures than
members from individualist cultures, such that collectivists would have more privacy concern than
individualists.
Note: Hypotheses 1 to 11 are tested in the first, survey-based part of empirical work. Additional,
follow-up hypotheses in response to the findings for H1 to H11 will be formulated for the second,
experiment-based part (see chapters 6 and 7).
2.5. Brief Outline of Research Overview
A theoretical framework, combining SIT, SCT and SIDE has been used to address the research
questions to evaluate the role of culture on the psychological variables (online group identity,
online self-enhancement, online perceived social support, online trust and online privacy concern)
46
on SNS. A cross-cultural data was collected by using UK (individualist country) and India &
Indonesia (collectivist countries) using an online survey and priming experiments. Study 1 and
Study 2 consists of the online survey while Study 3 and Study 4 consists of the priming
experiments. After the completion of the data, a quantitative analysis was conducted which will
be discussed in the later chapters. The findings of this study have theoretical, methodological and
practical implications which will also be discussed in later chapters. Figure 2.2 below is a visual
representation of the overview of the research which outlines how the research aim is guided by
the theoretical framework used in the study and attain results that has practical implications.
47
Figure 2.2.
Visual Illustration of Research Overview
Note: SNS = Social Networking Sites
48
Chapter Three
Methodology
3.1. Introduction
The previous chapter was an attempt to shed light on past research that has been conducted
which looked at each of the factors that are relevant for this thesis. The literature review brought
to light past findings and gaps that exists in in past studies. Keeping the research aim in mind, this
chapter will discuss the methodology that was used to answer the research aim and its hypotheses.
It will also discuss the various considerations that were taken into account to develop a consistent
research strategy cross-culturally. It is however important to point out that further in-depth
information on the methodology (which will discuss the sample, procedure and materials) in each
study will also be provided in the later chapters.
Note: section 3.1.4.3. Survey translation procedure is part of a published journal (Halder et al.,
2016).
3.2. Research Paradigm and Outline of Design
Cross-cultural studies systematically compare individual’s relationships developed and
behaviour in one culture with another culture. Although there has been an increase in cross-cultural
research since late 60s but has mostly been led by Western researchers. Berry (1989) further argues
that while it is difficult to remove the influence of one’s own culture when we are trying to
understand other cultures, researchers should make every effort to become submerged to
understand the new culture and he suggests that this could be achieved by looking at the constructs
in the perspective of cultural self-construal (individualism and collectivism), (see Hofstede, 1980).
Therefore, to understand the behaviour of participants on the psychological constructs a cross-
cultural method was implied. A quantitative approach was used to collect and analyse data which
used a survey design and a priming experiment. While the survey helped to understand the
relationship between the constructs, the experiment helped to understand the casual role of culture.
49
3.3. Research Population
Participants were recruited from UK, India and Indonesia and the sample consisted of a mixture
of students and participants who were employed too with an age range between 18 – 47 years,
justifying a wider demographic sample, which is hoped will help to address some of the sampling
issues in past research (Gardner, Gabriel, & Lee, 1999; Sedikides, Gaertner, & Toguchi, 2003),
however it is highlighted that even then the sample population was still dominated by student
participants in all three cultures. It is however argued that as there is a wider diversity of
participants with a wider age range, the results can be generalized to the overall population as even
student sample has diverse attitudes.
Around 30 percent of the population in India uses internet of which 23.1 percent uses Facebook
and around 30.5 percent of the population in Indonesia uses internet and 4.8 percent uses Facebook.
However, when compared to the rest of the world, India is the second highest user of internet after
China and Indonesia is the fourth highest internet user (Enrique, 2016). While there seems to be a
difference in Facebook use in India and Indonesia however it is also important to justify the reason
for using these two countries. While it was important to include countries, which could be
representative of Facebook use in the country, it was also important that the countries selected
were represented the cultural dimensions that was required for this study. While UK represented
the Individualist country, India and Indonesia represents collectivist countries.
India was the first choice as a collectivist country to be used for data collection for the study.
However, due to the delay in response from Indian educational establishments, Indonesia was also
chosen as the choice for collectivist country to recruit sample for the other studies as both India
and Indonesia were similar in the scores of collectivism and individualism attributes (Hofstede,
1980; Triandis et al., 1993). This helped in completing the data collection on time. While Study 3
(Experiment: Operationalization) was conducted in India, Study 1 (Survey: Operationalization,
Study 2 (Facebook responses) and Study 4 (Facebook and Priming responses) were conducted in
Indonesian.
50
3.4. Research Considerations
The below sections will outline the various considerations that were considered in order to
conduct a successful cross-cultural study.
3.4.1. Cultural consideration
The most important cultural consideration that had to be considered was the inclusion of both
male and female participants in the study as both India and Indonesia seemed to score high on
masculinity on Hofstede’s cultural dimension. To ensure consistency in the study, same procedure
for data collection was used in all countries. Care was taken that during data collection either the
researcher or a research assistant was present during the study to handle participant queries. In
addition, all the responses that were recorded in paper format was stored safely and destroyed once
the data was inputted in electronic format. All the data that has been recorded electronically has
been stored safely.
3.4.2. Language consideration
For the data that was collected in India and Indonesia, the research material was translated in
the native language of the country. For India, the survey was translated in Bengali (native language
of Kolkata) and in Bahasa for the Indonesian sample. In order to avoid errors and increase
equivalence across cultures a back-translation method was used (Behling & Law, 2000; Van
Widenfelt et al., 2005).
3.4.2.1. Survey Translation Procedure
The variability in different cultures makes it difficult to administer the same instrument
consistently in all cultures (Sechrest, Fay, & Zaidi, 1972; Sekaran, 1983). Only when we can use
the native language of the respondents which will ensure that they can fully comprehend the items
in the instrument which will then allow them to give their truthful responses. There needs to be a
balance between maintaining methodological requirement and literal meaning of the items
(Sechrest et al., 1972; Sekaran, 1983). Administering a culturally viable research instrument can
51
only be achieved when it is used in the participant’s native language which can be achieved by
translating the research instrument that will ensure the identification of such cultural variance
adequately. Although it has been argued that academic researchers can fail to identify individual
cultural differences if they focus too much on maintaining methodological equivalence, on the
contrary it can also be argued that only by maintaining such equivalence can we minimize the
variance in a data set. The actual approach needs to be the researcher’s individual decision. In the
present study, care was taken that participants in both cultures received the same information
throughout their participation as suggested by Sekaran (1983).
This particular study was in the format of a survey and very much text-based, the immediate
question was that of translation. One of the primary goals of translation is to obtain an instrument
that could be used consistently between cultures, i.e., with same literal and cultural meaning
(Brislin, 1970; McGorry, 2000). It is not only important to get at the correct “linguistic” translation,
but at the same time to achieve an equivalence in cultural meaning thereby helping the researcher’s
belief that intercultural differences in the findings are not due of translation error. There are
different ways of conducting a translation in a systematic way, e.g., the one-way translation
(without any back translation in the original language), translation by committee (this method
involves two or more individuals who are familiar with both languages who translate the
instrument from the original version. The researcher then uses the independent translators to arrive
at a consensus or can even recruit an additional independent party to choose the version that fits
best with the original version) and decentring (this method involves designing survey instruments
which fit in with the target culture and involves constant revision of the original survey instrument.
This method can also alter the items and survey length (McGorry, 2000). Using bilinguals, who
can read and write fluently in both languages are crucial during a translation process (Marin &
Marin, 1991). However, cultural phraseology should also be considered where regional or class
differences matter. So, it is first important to decide which groups of culturally different people
are under investigation.
Being able to develop a culturally and literally viable instrument was crucial for the research
design and for maintaining validity and reliability of the data. Therefore, two independent groups
of translators which consisted of three members in each group were recruited who were fluent in
reading and writing in both English and Bengali (the Indian native language most relevant in
52
Kolkata) in order to conduct the back-translation process adequately (see Figure 3.1). This process
of translation is recommended to be the most adequate (Marin & Marin, 1991) and was also cost
effective and quick unlike the decentring process of translation (McGorry, 2000) as it did not
depart from the original version of the survey instrument. It was also different to the translation by
a committee method since the three members in each group translated the survey items after a
group discussion and not as independent translators.
The translation method selected for this particular study had the advantage of quickly removing
any discrepancies in the translated version as survey items were translated after coming to a group
consensus. Although it is recommended that applications of several translation processes would
help us to achieve a more accurate and culturally fit instrument (McGorry, 2000), it is however
not always possible for researchers due to restrictions in funding and time constraints. Below is
the step by step guide of the back-translation process used.
3.4.2.1.1. Back Translation Procedure for research instrument in India
1) A focus group of three Bengali translators were selected who have studied and lived in
Kolkata. They could read, write and speak fluently in Bengali and in English.
2) The focus group was asked to read through the questionnaire which was in the original
English version and then after discussion with each other came up with a Bengali version of each
item on the survey.
3) Another focus group of three Bengali translators were selected for the back translation
procedure. They could also read, write and speak fluently in Bengali and English.
4) The second focus group was asked to read every statement in the Bengali questionnaire
and come up with an English version of each statement on the questionnaire. This was also
achieved with discussion between the three members in the focus group.
5) The original English version of the questionnaire was then compared with the second
English version of the questionnaire by all parties including the researcher. Although both the
original English version and the second English version of the questionnaire were not exactly the
same, the core meaning of each survey item was found to be well preserved. Hence there were no
53
further changes made to the translated version of the questionnaire for the study. On completion
of the translation process the survey were administered to participants in India.
The below process outlines the translation process that was followed in India followed by Figure
3.1. which illustrates the overall back translation process that was used throughout.
3.4.2.1.2. Back Translation Procedure for research instrument in India
1) A group of back translators were recruited who helped to translate the research material in Bahsa
Indonesia (native language in Indonesia). The translators were undergraduate students in Indonesia
who could speak, read and write in Bahsa Indonesia and English and lived in Indonesia.
2) At first a focus group of two English and Bahsa Indonesia fluent translators read through
the English version of the questionnaires and after discussion with each other translated each item
on the questionnaires in Bahsa Indonesia version.
3) The Bahsa Indonesia versions of the questionnaires were then back translated by another
focus group of two who translated back each item into English. The two back translators were
researchers and was pursuing a PhD from a UK university. They were both lecturers in reputed
Indonesian universities. Both back translators were could speak, read and write in English and in
Bahsa Indonesia.
4) The original English version of the questionnaire was then compared with the second
English version of the questionnaire by all parties including the researcher (see Figure 2 for the
full translation process). During instances when the back translators did not agree with any
particular word used in Bahsa language, the item in the Bahsa Indonesia version of the
questionnaire was changed to fit in with the meaning of the survey item after careful discussion
with each other. Care was taken to make sure that the core meaning of the item was preserved.
5) Once all the items had been checked, the changes in the Bahsa Indonesia version was made and
the final version of the translated questionnaire was then scanned and emailed to the research
assistants in Indonesia.
On completion of the translation process the surveys were circulated among the participants in
Indonesia. Figure 3.1 below illustrates the overall back translation process followed in both
countries.
54
Figure 3.1.
An Illustration of the Back Translation Process followed in India & Indonesia
3.4.3. Ethical considerations
Throughout the process of the study all guidelines outlined by the British Psychological Society
and the Nottingham Trent University were adhered to. Prior to the data collection, ethical approval
was gained from the Nottingham Trent University’s Ethics Board following the submission of all
the research questions, instruments, participant consent and debrief forms (see Appendix). As it
was a cross-cultural survey clear information on the data collection process was also outlined. Due
to the cultural differences involved in the study, all participants were clearly informed of what
would be expected from them and also about their rights. The participants were clearly informed
of that their participation was completely voluntary and that all their data would be kept
anonymous and would be identified with their unique number only.
55
The data collected was stored in a secured drive and were identified by unique numbers only.
All paper copies of the responses were destroyed after the data had been saved online on the Bristol
Online Survey repository both for the UK and Indonesian sample. For the Indian sample, all data
once saved on a secured drive the paper copies of the responses were destroyed. As a thank you
for their participation all participants had the chance to enter in a lottery draw to win prizes which
included vouchers from their respective countries. The winners in all the countries were emailed
to inform that they had won the lucky draw and the prizes were sent to them by post. No monetary
award was given to the participants.
Although this cross-cultural research was conducted taking into considerations all ethical
considerations, however there were several challenges that were encountered during data
collection and in India (see Appendix 13). This section outlines the various challenges faced during
the data collection in India with an aim of helping future researchers.
3.5. Overview of Study Design
The data collection was conducted in India, Indonesia and in the UK. All the surveys were
developed keeping in mind the research questions. This was achieved by thorough in depth
literature review and thorough discussion with the supervisory team. Below is an overview of the
four studies that was conducted during the research process (see Table 3.1).
a) Study 1: Survey (Operationalization) – An online survey was conducted in UK and
Indonesia. While the survey in UK was conducted online, the translated version of the
survey was circulated in paper format for the Indonesian participants and their responses
was then submitted by a research assistant online in English. The study was further refined
after careful considerations of the feedback received from Study 1.
b) Study 2: Survey (Facebook responses) – This online survey was a refined version of
Study 1 and was circulated among the UK and the Indonesian sample. The Indonesian
sample were given the translated version of the questionnaire which was later submitted
online by two research assistants.
c) Study 3: Experiment (Operationalization) – Surveys in the form of questionnaires which
had two versions was conducted in UK and in India. While the questionnaire was
56
conducted in English for the UK participants, a translated version of the questionnaire was
circulated among the Indian sample. All the responses were submitted online by research
assistants.
d) Study 4: Experiment (Priming & Facebook responses) - This was the refined version of
the experiment study (Study 3) and was circulated among the UK participants and the
Indonesian sample. The Indonesian sample were given the translated version of the
questionnaire which was later submitted online by research assistants.
Data analysis was conducted after each study using SPSS version 22 and 23. During this process
all feedback received from participants during were taken into consideration which helped in the
development of amended versions of the instruments (both survey and experiment). The amended
versions of the instruments had in depth and clearer instructions and had additional scales which
helped to answer research questions.
Table 3.1
An Illustration of the Study Design for the Thesis
Study Type Country
Study 1 Survey (Operationalization) UK & India
Study 2 Survey (Facebook responses) UK & Indonesia
Study 3 Experiment (Operationalization) UK & Indonesia
Study 4 Experiment (Priming & Facebook responses) UK & Indonesia
3.6. Summary
This chapter was an attempt to provide an outline of the methodological considerations and
bring to light the studies conducted. Attempt was also made to highlight the importance of cross-
cultural study and discussed how the translation of research instruments are crucial in capturing
valid responses from research participants. This chapter also highlighted the entire process of data
collection through four different studies (Study 2, Study 3 and Study 4). The next chapter four
57
chapters are the Results section which will discuss each Study conducted and analysed in further
detail.
58
Chapter Four
Study 1: Survey (Operationalization)
4.1. Rationale for this study
The main aim of Study 1 is to use it as a testing ground to check whether the data collection
process, survey design could consistently be used in both countries successfully. It was also further
aimed to check if the measuring scales used in the study was best suited to answer the research
aim which was to evaluate the role of culture on various psychological variables in an online
environment. While this study helped to conduct preliminary analysis of the data, further detailed
analysis will be discussed in Study 2. While it is known that culture does influence behaviour, it
envisaged that such cultural differences also exist in an online environment thereby helping us to
understand its impact on some of our psychological variables which effects how members behave
on SNS. In order to clearly evaluate any cultural effect this study was conducted in Indonesia (a
collectivist country) and in UK (an individualist country).
The scales in the study consisted of several constructs which looked at online behaviour like
cultural self-construal (which measured cultural orientation of the participants), social support,
group identity, trust, self-enhancement and privacy. While the cultural self-construal scale
consisted of survey items that measured overall cultural orientation and included items measuring
individualist traits and collectivist traits, the other items were measured in relation to the online
environment. For example, social support (items measured how to what extent participants
received social support online, group identity (items measured how much members identified with
their social network), trust (items measured how much participants trusted their social group on
Facebook, self-enhancement (items measured how much participants self-enhanced on Facebook
when compared with others and privacy (items measured how much participants were concerned
about their online privacy). Most of the scales were validated scales adapted from past literatures
and will be discussed in the below paragraphs. Once the scales were finalized, the format of the
questionnaire was decided. The questionnaires were designed keeping in mind that the
questionnaires would be circulated in both western and eastern cultures. Due to this the language
was kept simple and translated versions of the questionnaires was used for India and Indonesia as
59
English was not their first language. After the data was collected all data was analysed using SPSS
version 22 and 23 and other advanced statistical packages like PROCESS and AMOS.
4.2. Method
4.2.1. Participants
A total of 150 (UK: N = 47; 76.6 per cent female; Indonesia: N = 103; 65 per cent female)
university students in the UK and Indonesia participated in a survey and were all Facebook users.
The mean age of the participants was UK: 20.51 years (SD = 5.02) and Indonesia: 20.54 years (SD
= 2.37) with 97.9 per cent of the respondents falling in the range from 18 to 27 years.
4.2.2. Procedure
An online survey was developed on the Bristol Online Survey (BOS) website and was
administered across university students in UK. For the Indonesian sample, an Indonesian research
assistant was recruited who was fluent in both English and their native language who helped to
circulate the paper version of the survey to the University students in Indonesia and then inputted
the responses online on the BOS in English. The data input on the BOS website was also double
checked by another research assistant who was also recruited to avoid any mistakes.
While UK participants received the English version of the survey the Indonesian participants
were given the translated version. The survey lasted for approximately 15-20 minutes. Full
participant consent was attained and participants were also made aware of their rights before their
participation in the survey. Participants were also given the chance to enter in a lottery draw to win
gift voucher as a thank you for their participation which was completely voluntary.
4.2.3. Materials & Description of items
Study 1 was a survey development study, where participants were given full participant
information about the survey and their rights. It also had information on the chance to enter in a
lottery draw after the survey which was completely voluntary. After completion of the survey
60
participants were debriefed and winners of the lottery draw were notified with their gift vouchers
in both the country. The survey collected demographic information like “gender”, “age”,
“ethnicity”, “work: employed –full time; employed – part time; unemployed; student – full time;
student – part time”. It also checked the “Country of Residence”, “Nationality”, and “First
language” of the participants. Participants were also asked to enter a unique code (combination of
number and letters) which was to be used to identity their survey in case of any queries or issues.
In order to take part in the survey participants had to be members on Facebook as some of the
questions in the survey was in relation to their Facebook usage.
In order to understand network diversity, participants were asked to provide the number of
members (to the best of their knowledge) they knew in each category both offline and online. The
categories included were “Immediate family”, “Other birth family”, “Family of spouse or
significant other”, “Co-workers”, “People at work but don’t work with directly”, “Best
friends/confidantes”, “People known through hobbies or recreation”, “People from religious
organizations”, “School relations”, “Neighbours”, “Just friends”, “People known through others”,
“People who provide a service” and “Any other”. Participants were informed that people they
know under each category could also be known to them online and offline. The intent of this was
to evaluate the network size participants have online and offline. The overall network size for each
category was then computed for both online and offline network.
In order to understand members, use of Facebook participants were asked to indicate the
average number of hours (in minutes) and average number of days they used Facebook in the last
fourteen days. The minutes were converted to hour for final analysis and an overall mean score
was computed for the days’ participants used Facebook “FB Days” and the amount of time they
invested while on Facebook “FB HR”. Participants were then asked to think about the use of
Facebook in the last fourteen days and indicate on average in minutes their usage during weekday
and weekend for “study – used Facebook solely for study purpose”, “work – used Facebook solely
for work purpose”, “social life – used Facebook just for socializing” and “mixed – used Facebook
for work/ study and socializing at the same time”. The minutes was converted to hour for final
analysis. Participants were then asked to also rate on several scales adapted from past research.
The below sections will provide details of the scales that were used in the study.
61
4.2.3.1. Cultural Self-Construal scale: This scale measured the level of cultural self-construal of
the participants and has been adapted from validated scales used by Singelis (1994). The original
scale had 12 items for measuring collectivism self-construal and 12 items measuring individualism
self-construal. For this study 8 items measuring collectivism self-construal and 8 items measuring
individualism self-construal scale was used in order to fit in the most revenant items for this study.
Some of the wordings of the scale was changed to fit in with the current study. For example, one
of the item in the independent scale was “Speaking up during a class is not a problem for me” was
changed to “Speaking up is not a problem for me”. Also, another item in the independent scale
was “I am the same person at home that I am at school” was also changed to “I am the same person
at home that I am during social gathering”. The cultural self-construal scale was divided into two
separate self-construal scales, one measured the level of individualism for the participants and the
other measured the level of collectivism for the participants. Respondents were asked to rate the
items in relation to their Facebook contacts and also in relation to their overall social contacts (this
included both online and offline social contacts they knew). The items were coded using five point
Likert scale ranging from “Strongly Disagree” to “Strongly Agree”. For this study “Strongly
Disagree” was coded as 1, “Disagree” was coded as 2, “Neither agree or disagree” was coded as
3, “Agree” was coded as 4 and “Strongly Agree” was coded as 5. For a full list of the items see
Appendix 1. The internal consistency of the collectivism measuring items was (Cronbach alpha =
.61) and for the individualism measuring items was (Cronbach alpha = .56). In order to calculate
the overall cultural orientation (CO) score for the participants, the items measuring the level of
individualism self-construal reversed scored so that it is in line with the collectivism measuring
items as higher score on the collectivism scale would mean lower score on individualist traits. The
same process was followed for both the Facebook cultural orientation and also overall cultural
orientation of the participants.
4.2.3.2. Facebook Activity Scale: This scale was included to evaluate the different types of
activities participants use when they are on Facebook. This scale adapted from past validated scale
used by Junco (2012) and had a scale reliability of (Cronbach alpha = .88). Respondents were
asked ‘how often (on average) have they engaged in the following activities on Facebook in the
last 14 days’. The respondents were asked to give their best rating. Facebook activity was coded
using five point Likert scale ranging from “Never” to “Very often (close to 100% of the time)”.
For this study “Never was coded as 1, “Rarely” was coded as 2 (25%), “Sometimes” was coded as
62
3(50%), “Often” was coded as 4 (75%) and “Very often” was coded as 5 (100%). Respondents
were asked to rate as per the best of their knowledge how often (on average) they involved in such
Facebook activities. An overall mean score for the scale was also computed.
4.2.3.3. Online Group Identification Scale: The group identification scales looked at how much
respondents identified with their social group on Facebook and was adapted from past validated
scale used by Brown et al. (1986) however only six out of the original ten items were used in the
study. Some of the phrasing of the items in the scale were also changed to fit in with the study. For
example, the item “I am the person who identifies with the group” was changed to “I see myself
as a member of my Facebook community”, “I am a person who considers the group important”
was changed to “It is important for me to be a member of my Facebook community”, “I am a
person who is glad to belong to the group” was changed to “I like being a member of my Facebook
community” and were the three positively rated item. The other three reversed items were used in
the scale to get the truthful responses from the respondents and avoid any false information as
these reversed items will force respondents to think and provide their ratings. Again, the item “I
am a person who makes excuses for belonging to the group” was changed to “I am not glad to be
a member of my Facebook community”, the item “I am a person who tries to hide belonging to
the group” was changed to “I am not proud to be a member of my Facebook community” and the
item “I am a person who feels held back by the group” was changed to “I do not like being a
member of my Facebook community”. The negative items were phrased similarly to enhance
participant engagement. The internal consistency of the amended version of the six items in the
group identification scale was high (Cronbach alpha = .75) suggesting that the items measured the
same construct. For this scale, too an overall mean score for the scale was computed for analysis.
4.2.3.4. Online Self-Enhancement Scale: This scale was designed as an attempt to show that self
enhancement does exist in both Western and Eastern cultural and had been adapted which has been
validated in the past (Gaertner, Sedikides, & Chang, 2008; Sedikides, Gaertner, & Toguchi, 2003).
Participants were asked to provide their responses on a six point Likert scale ranging from 0
(lowest score) to 5 (highest score) The higher the rating the higher they score on that attribute.
They were first asked to rate themselves on the cultural traits, then their Facebook contacts
followed by their rating for their Offline social contacts (this could also include members that
might also be on their Facebook contacts). In order to find out if respondents rated themselves
63
higher than their social contacts on Facebook, the rating they provided for each category for
themselves was subtracted from the ratings they provide for their social contacts on Facebook. For
example, if the respondent had rated “Independent” category as “5” for themselves and “3” for
their social contacts on Facebook then “5-3 = 2” was the difference which was considered as the
self enhancement score for that participant for that particular category, i.e., they had rated
themselves higher on that category. In this way all the individual category was calculated by using
the Compute function in SPSS and an overall rating for each category was calculated at the end.
The difference in the rating which resulted from this would was taken as the self enhancement
score for that particular category. It is to be noted that for this study analysis was only conducted
on the ratings respondents gave for themselves and their Facebook social contacts. The internal
consistency suggests a high Cronbach alpha of .64 for the collectivism measuring items and
moderate Cronbach alpha of .54 for the individualism measuring items. An overall mean score for
the collectivist and individualist attributes were computed separately.
4.2.3.5. Perceived Online Social Support Scale: This scale was included to evaluate how often
participants receive social support online. This was a validated scale and was adapted from
Sherbourne & Stewart (1991). Respondents were asked “to think about the people they know on
Facebook and were asked to rate how often they have received the type of social support listed in
the scale on Facebook”. The items were coded using the Five point Likert scale ranging from
“None of the time” to “All of the time”. For this study “None of the time” was coded as 1, “A little
of the time” was coded as 2, “Some of the time” was coded as 3, “Most of the time” was coded as
4 and “All of the time” was coded as 5. The MOS scale was divided into three sub scales, i.e.,
emotional/ informational support, tangible support, affectionate support, positive social interaction
and an additional item. However, for this study only items that were relevant to the online world
were used from the social support scale. For example, one of the item under the Tangible sub scale
was “Someone to help you if you were confined to a bed” and “Someone to help you with your
daily chores if you were sick” are some of the examples of items that did not fit in with this
particular study as it might not be relevant in an online environment. Therefore, items that did not
fit in with the study was excluded for our study. Internal consistency of the items was very high
with a Cronbach alpha of .96 which confirms that the items measures the same construct. An
overall mean score of the ratings was computed for the scale.
64
4.2.3.6. Online Trust Scale: This scale measured the level of trust that participant had for their
social contacts on Facebook and was adapted from past validated scale used by Krasnova et al.
(2010). The items were coded using five point Likert scale ranging from “Strongly Disagree” to
“Strongly Agree”. For this study “Strongly Disagree” was coded as 1, “Disagree” was coded as 2,
“Neither agree or disagree” was coded as 3, “Agree” was coded as 4 and “Strongly Agree” was
coded as 5. The items in the scale were a) Do their best to help me, b) Do care about the wellbeing
of others, c) Are open and receptive to the needs of each other, d) Are honest in dealing with each
other, e) Keep their promises and f) Are trustworthy. The internal consistency of the items was
high and had a Cronbach alpha of .88. An overall mean score for the scale was computed as well.
4.2.3.7. Online Privacy Concern Scale: This scale was developed to check the level of privacy
concern on Facebook among the participants after thorough discussion in a focus group of three
members who were also researchers in the University in UK. The items were designed to fit in
with the current study and care was taken to come up with items that measured the construct. The
scale had seven items in total (five positively worded and two negatively worded items). The items
were presented in the survey in a counterbalanced way in order to enhance participant engagement.
Respondents were asked to give their rating on a five point Likert scale ranging from “Strongly
Disagree” coded as “1” to “Strongly Agree” coded as “5” where “1” was the lowest rating and “5”
was the highest rating. One of the negative item “I don’t care who looks on my Facebook profile”
was removed and scale reliability increased from .57 to .70. The final set of items consisted of 5
positively worded items and 1 negative item that was reversed scored. An overall mean score was
computed for this scale too.
The below section will discuss the results and the analysis followed by a brief summary. This
will then follow with a section which highlights the various amendments made after Study 1 and
then will discuss Study 2. Finally, an overall summary of the findings will be presented.
4.3. Results and Analysis
This section attempts to evaluate the relationship between the independent variable (IV)
cultural orientation and dependent variables (DV) which are (online group identity, online self-
enhancement, perceived online social support, online trust and online privacy). It aims to evaluate
65
to what extent use of Facebook (through frequency and type of activities) has an influence on the
DVs.. The section is divided into seven sections. The first section will look at sample distribution
through descriptive statistics. The next section is aimed at evaluating the relationships of the
variables through bivariate correlations, multivariate relationships using SPSS version 22 and
mediation analysis using statistical technique called Process and then the models are constructed
and tested using Structural Equation Modelling (SEM) using AMOS where the theoretical validity
of relationships of the variables are tested. SEM is used to test the relationship between the
hypothesized model and the observed data. Model fit is attained in AMOS which helps to support
the findings. Section five looks at the limitations of the models and proposes plans for future
analyses.
4.3.1. Descriptive statistics
Descriptive statistics is first used to test individual variable through frequency distribution and
explore options in SPSS which checks whether further in depth analysis should be carried out.
Frequency tables first explores the number of respondents in the data, their frequency counts. The
explore option looks at the missing data and outliers and also looks at the kurtosis and skewness
of the data.
4.3.2. Data screening and preliminary analysis
Data was analysed using SPSS version 22. The threshold for significance was set to α = 0.05
for all decisions. Prior to in depth analysis data was first screened to check for missing data and
outliers. Boxplots and subsequent checks revealed the presence of some outliers and missing data.
All outliers and missing data were replaced with the mean for the variables. The assumptions of
normality were also checked if there was any presence of skewness or kurtosis in the data.
Normality of the data was accepted if the distribution of the data, i.e., both skewness and kurtosis
was within range (z < 2.58) (Hair et al., 2006).
4.3.3. Sample characteristics
66
Table 4.1. Demonstrates the characteristics of the participants. The overall mean of the UK
respondents was 20.51 (SD = 5.02) and the age range were between 18 to 47 years. The mean age
of the Indonesian respondents was 20.54 (SD = 2.37) years. The age range for the Indonesian
respondents were 17 to 27 years. The age range for the UK respondents were more varied than the
Indonesian sample (see Table 4.1) however overall sample consisted of more than 90 percent of
student population (full time and part time students included) and more than 80 percent in the
Indonesian sample (see Table .42). The highest number of respondents were from the population
from 18 – 21 years of range for the Indonesian sample and between 18 – 20 years’ age range in
the UK sample.
Table 4.1.
Demonstrates the descriptive for both UK and Indonesian sample
Variable Mean (SD)/ Frequency Range/ Percentage
Overall (N= 150)
Age (years):
UK 20.51 (5.02) 18 - 47
Indonesia 20.54 (2.37) 17 - 27
Gender:
UK 1.77 (.43) -
Indonesia 1.65 (.48) -
Country: UK: N = 47; Indonesia: N = 103)
Gender:
UK : Male 11 23.4
: Female 36 76.6
67
Indonesia : Male 36 35.0
: Female 67 65.0
Note: Overall represents both UK and Indonesia combined
Table 4.2.
Demonstrates Ethnicity and Work characteristics in UK and Indonesia
Variable Mean (SD)/ Frequency Percentage
Overall (N= 150)
Ethnicity:
UK 47
Indonesia 103
Work:
UK: (N = 47)
Employed (full time) - -
Employed (part time) 3 6.4
Unemployed - -
Student (part time) - -
Student (full time) 44 93.6
Indonesia: (N = 223)
Employed (full time) 4 3.9
68
Employed (part time) 2 1.9
Unemployed 6 5.8
Student (part time) 6 5.8
Student (full time) 85 82.5
Note: Overall represents both UK and Indonesia combined
Table 4.3. Below demonstrates mean use of Facebook (in hours) during the weekdays and
weekend. The data demonstrates that Facebook is mostly used for socializing in both UK (weekday
M = 3.11 and weekend M = 2.49) and in Indonesia (weekday M = .62 and weekend M = .86). An
independent sample t-test revealed that there was a significant difference between countries when
participants used Facebook for ‘study during weekday in UK (M = 1.54, SD = 5.07) and in
Indonesia (M = .26, SD = .46), t(148)= 2.93, p = .004, suggesting that participants in the UK used
Facebook for study purpose more than Indonesian participants in the weekday. However no
significant difference was found for Facebook use during the weekend (M = .72, SD = 2.18),
t(148)= -1.05, p > 0.05. There was also a significant difference found between use of Facebook
for social purpose during the weekday by participants in the UK (M = 3.11; SD = 5.94) than
participants in Indonesia (M = .62; SD = .90), t(148) = 6.24, p = .001 and also during the weekend
(UK: M = 2.49; 2.86) and (Indonesia: M = .86; SD = 1.89), t(148) = 5.19, p = .001. There was also
a significantly higher use of Facebook for mixed purposes (which included study, work and social)
during the weekday by participants in the UK (M = 2.53; SD = 6.05) than by the participants in
Indonesia (M = .79; SD = 1.30), t(148) = 2.54, p = .012. Whereas for all the other categories the
differences in Facebook was non-significant. This overall suggest that participants in the UK use
FB more than participants in Indonesia and both countries seem to be using FB more during the
weekdays and mostly for social purposes.
69
Table 4.3:
Demonstrates FB use per weekday & per weekend in UK & Indonesia (in hours)
UK (N = 47) Indonesia (N = 103)
Weekday Study Work Social Mixed Study Work Social Mixed
Mean 1.54 .32 3.11 2.53 .36 .26 .62 .79
SD 5.07 .73 5.94 6.05 .56 .46 .90 1.30
Weekend Study Work Social Mixed Study Work Social Mixed
Mean .72 .32 2.49 2.00 .57 .28 .86 .97
SD 2.18 .98 2.86 5.64 .96 .49 1.19 1.57
Note: FB = Facebook
A t-test conducted further suggests that there was a significant difference in the number of days
members used FB between the two countries (M = 12.47; SD = 2.84) and Indonesia (M = 6.14;
SD = 4.58), t(148) = 8.73, p = .001) suggesting that participants in the UK tends to use Facebook
more than Indonesian participants. However there no significant difference found between the
countries on the number of hours they invested on Facebook (UK: M = 2.77; SD = 8.05; Indonesia:
M = 3.63; SD = 7.49), t(148) = -.643, p > 0.05). Table 4.4. below illustrates that on average UK
participants spend at least 4 days out of the 14 days on Facebook whereas the Indonesian sample
seem to suggest that there can be occasion where they don’t use Facebook at all. This suggests that
the use of SNS like Facebook is more popular among the UK participants than the Indonesian
participants.
70
Table 4.4.
Illustrates Frequency of Facebook use among UK & Indonesian sample
UK
(N = 47)
Indonesia
(N = 103)
Overall
(N = 150)
Number of Days Facebook used in the last 14 days
Mean 12.47 6.14 8.12
SD 2.84 4.58 5.05
Minimum 4 0 0
Maximum 14 14 14
Average time (hours) spent on Facebook in the last
14 days
Mean 2.77 3.63 3.36
SD 8.05 7.49 7.65
Minimum 1 0 0
Maximum 56 56 56
Scale reliability of the items was also measured. While the majority of the items had good to
high scale reliability, the scale measuring the cultural orientation of the participants had a medium
scale reliability (see Table 4.5.). However, the low scale reliability could be attributed to the small
sample size of the study. Apart from the privacy scale which still achieved a moderate scale
71
reliability (Cronbach alpha = .61), all the other scales had been validated in past research studies,
hence it was decided to continue the data collection using the below scales.
Table 4.5.
Reliability scores of the variables used in the Survey
Variable Cronbach alpha [α] Mean SD Number of
items
1) Facebook activity .88 33.59 8.89 14
2) Social Support .96 30.45 9.78 10
3) Trust .88 19.45 3.82 6
4) Privacy .61 18.83 3.34 6
5) Group Identity .75 19.29 2.87 6
6) Self enhancement
a) Individualism categories .70 23.70 3.81 7
b) Collectivism categories .56 25.43 3.42 7
7) Cultural Orientation
a) Individualism .54 29.35 3.03 8
b) Collectivism .61 28.51 3.16 8
A Pearson’s bivariate correlation was conducted using SPSS to check the relationship between
the items for each construct as it would help to decide if further analysis should be carried out (see
Table 4.6) suggesting that there was a significant positive correlation between online trust and
online privacy (r = .161, n = 150, p < 0.05) suggesting that as members increase their level of trust
72
online their level of privacy concern also seems to increase. There was also a significant positive
correlation found between online trust and group identity (r = .302, n = 150, p < 0.01) suggesting
that as the level of trust increases on Facebook, identification with their group members also
increases. There was a significant negative correlation between self enhancement and online trust
(r = -.209, n = 150, p < 0.05). There was a significant positive correlation found between social
support received online and online privacy concern (r = .256, n = 150, p < 0.01) suggesting that
increase in online privacy helps to increase online social support. There was also a significant
positive correlation found between Facebook activity and group identity (r = .209, n = 150, p <
0.01) and also a significant positive correlation between Facebook activity and online social
support (r = .161, n = 150, p < 0.05) suggesting that increase in Facebook activity helps to increase
group identity and also help to receive online social support. There was a significant positive
correlation found between country difference and group identity (r = .219, n = 150, p < 0.01); a
significant negative correlation between country difference between country difference and self
enhancement (r = -.253, n = 150, p < 0.01), social support (r = -.489, n = 150, p < 0.01) and
Facebook activity (r = -.432, n = 150, p <0.01). Indonesia was coded as 2 and UK was coded as 1.
Country was coded as (UK = 1 and Indonesia = 2), therefore the results suggest that increase in
country, would increase identification with their social group. Indonesian social groups are mostly
made of contacts that they might share an offline relationship; therefore they might be connected
to their group members more than their UK counterparts. Increase in country code decreased self
enhancement, social support one receives online and Facebook activities. Therefore, in accordance
with the coding in the study, the results could be interpreted as Indonesian sample might be
reluctant to seek social support for the fear of being identified among their offline social group and
thus fail to self-enhance. The decrease in Facebook activity could also be related to the availability
of Facebook usage due to their commitments and societal restrictions.
There was a significant negative correlation found between age and social support (r = -.256, n
= 150, p < 0.01) and cultural orientation (r = -.187, n = 150, p < 0.05). For this study individualism
was coded as 1 and collectivism was coded as 2. This suggest that as age increases social support
decreases. This could be explained that with age members might develop established social
contacts offline from who they prefer to receive social support than from members who are online.
Additional the data also suggests that increase in age seems to decrease the cultural traits online as
members might become more accustomed to the norms and behaviours members follow on
73
Facebook. There was also a significant positive correlation between gender and online social
support (r = .164, n = 150, p < 0.05). For this study male was coded as 1 and female was coded as
2, therefore suggesting that female seem to receive more social support online than male
participants. It could be argued that female participants are more open in discussing their concerns
than male participants.
Table 4.6.
Illustrates correlations between variables (unstandardized items)
Privacy ID Trust SE SS CO Activity Country Age Gender
Privacy 1
ID -.002 1
Trust .161* .302** 1
SE -.019 -.103 -.209* 1
SS .256** -.058 .067 .157 1
CO .029 .043 .074 .054 .133 1
Activity .096 .209* .074 .119 .357** .088 1
Country .078 .219** .160 -.253** -.489** -.028 -.432** 1
Age -.107 -.128 -.100 .053 -.256** -.187* -.158 .005 1
Gender .086 -.058 -.018 .111 .164* -.096 .009 -.115 -.012 1
Note: ID = group identity; SE = self-enhancement; SS = social support; CO = cultural orientation
*p<0.05 (two tailed); **p<0.01 (two tailed)
Indonesian participants identified with their group members online (M = .15) more than the UK
participants (M = -.32). The mean difference between the samples was -.470 and the 95%
74
confidence interval for the estimated population mean difference was between -.810 and -.129.
The effect size was large (d = .45). An independent sample t-test showed that the difference
between the sample was significant (t = -2.726, df = 148, p = .007, two tailed).
A t-test was conducted to check the differences in the samples on the variables used. The results
suggest that for the self enhancement scale UK participants seemed to self-enhance more (M =
.37) on when they are on Facebook than Indonesian participants (M = -.17). The mean difference
between the samples was .543 and the 95% confidence interval for the estimated population mean
difference was between .234 and .851. The effect size was large (d = .58). An independent sample
t-test showed that the difference between the sample was significant (t = 3.486, df = 148, p = .001,
two tailed).
Indonesian participants trust their social group more (M = .11) than the UK participants (M = -
.24). The mean difference between the samples was -.343 and the 95% confidence interval for the
estimated population mean difference was between -.687 and .002. The effect size was large (d =
.34). An independent sample t-test showed that the difference between the sample was significant
(t = -1.967, df = 148, p = .05, two tailed).
UK participants received more social support (M = .72) when they are online than the
Indonesian sample (M = -.33). The mean difference between the samples was 1.051 and the 95%
confidence interval for the estimated population mean difference was between .746 and 1.355. The
effect size was large (d = 1.142). An independent sample t-test showed that the difference between
the sample was significant (t = 6.818, df = 148, p = .001, two tailed).
UK participants used Facebook more (M = .64) than the Indonesian participants (M = -.29).
The mean difference between the samples was .929 and the 95% confidence interval for the
estimated population mean difference was between .615 and 1.244. The effect size was large (d =
1.035). An independent sample t-test showed that the difference between the sample was
significant (t = 5.835, df = 148, p = .001, two tailed).
The difference in means between the UK and Indonesian samples was not significant for online
privacy concern (UK: M = -.11; Indonesian: M = .05), t = -.817, df = 148, p > 0.05) and for
measures of cultural orientation (UK: M = .04; Indonesia: M = -.02, t = .279, df = 148, p > 0.05).
75
Although differences in the two countries were established in the various psychological
variables used which suggest that there exists correlation between the constructs (see Table 4.6),
however it was important to find out how members scored on each of the psychological construct
based on the scores on the other remaining constructs. As the study involves human participants,
being able to single out each predictor variable was impossible as the impact of the constructs co-
existed. Unlike bivariate correlation which did not help to evaluate the causal relationship of
variables, multiple regression allowed the use of several predictor variables (DVs) which together
allow to estimate a participant’s likely score on a criterion variable (IV) (see Brace, Kemp, Snelgar
R., 2009). It was for this reason that multiple regression was used as an additional step to
understand participant scores on each psychological variable. Furthermore, as this was a cross-
cultural study the data collected was based on same measuring scales as such analysing both
countries in one data set was required as it would allow to evaluate country differences. By being
able to predict Facebook activities helped to check for correlations relevant to the theory.
In order to check the impact of culture on each psychological variable a multiple regression
analysis using the ‘Enter’ method was used to test if all the factors in the study for online group
identity, online self-enhancement, perceived online social support, online trust and online privacy.
All the assumptions for regression analysis were including multi-collinearity between IV and DVs
which were met. The interaction effect by country was also evaluated for each psychological
variable used. For example, country x online group identity, country x self-enhancement, country
x perceived online social support, country x trust and country x online privacy concern. The results
indicated that the overall model explained 15% of the variance (R2 = .23, F(12,137) = 2.90, p =
.01). It was found that only online self-enhancement predicted Facebook activities (β = .86, p
=.013) and there was also a significant negative interaction between country x self-enhancement
(β = -.43, p = .024). Age and gender was also used as predictors in the model as control variables.
As it was a cross-cultural design the impact of differences in countries used was also evaluated
through interactions. The results suggest that there was a significant negative interaction between
country and self-enhancement (β =-.77, p=.02) (see Table 4.7 for full information).
Table 4.7.
Illustrating the Standardized Regression Coefficients predicting Facebook Activities
76
B SE β
Self enhancement .86 .34 .86*
Privacy -.14 .25 -.14
Group Identity .34 .27 .34
Trust -.13 .28 -.13
Social support .20 .30 .20
Cultural orientation .23 .24 .23
Country x self-enhancement -.43 .19 -.77*
Country x privacy .10 .17 .15
Country x group identity -.05 .16 -.08
Country x trust .06 .17 .10
Country x social support .07 .19 .12
Country x cultural orientation -.14 .16 -.21
Gender -.07 .17 -.03
Age -.01 .02 -.04
Adjusted R2 .15
F 2.89
Note: *p<.05
4.3.4. Analysis
Study 1 was a survey that checked differences in Facebook responses. The results indicated that
there exists difference in the time participants invest when they interact on social media which is
influenced by country difference. However, both countries seem to use Facebook mainly for social
77
purposes which was expected. The results also indicated that participants in the UK engaged in
communicating through SNS more than Indonesian sample further suggesting the importance of
country difference. However, this might also be influenced by other factors such as availability of
the internet, social restrictions or even economical situations. Use of Facebook is seen to be helping
in positive self enhancement such that increase in Facebook activity increases feeling of self-
enhancement in both cultures however such effect cannot be discussed on its own as the results
also highlighted the influence of country difference. Increase in age seemed to decrease social
support online. This could be because with age members establish social network offline from
whom they prefer seeking social support as they might develop a more trusting relationship and
thus refrain from seeking social support online. The results also indicated that female participants
received more social support than their male counterparts suggest that being a part of the online
network helps to develop more opportunity for accessing social support for its members. However,
it could be argued that this result could be influenced by the higher number of female participants.
The findings achieved in this study will be explored further in Study 2.
4.4. Brief summary of Study 1
The main aim of Study 1 was to evaluate the cultural influence on the psychological variables
that has been used in the study. The results overall suggest that there exist differences in behaviour
between members from UK and Indonesia influenced by the differences in countries members
came from. Clearly online self-enhancement is seen to be influenced due to country differences.
While the other psychological variables failed to show any significant differences, it could be due
to sample size. Therefore, this certainly laid the foundation to acknowledge that there exist
differences in behaviour in an online environment. Therefore, in order to evaluate differences in
behaviour due to cultural orientation, a further study with a representable sample size and an
amended version of the survey needed to be designed. The below section will discuss the various
changes that were made in the survey which helped to define Study 2. Although these changes
were not part of the initial planning process, however as Study 1 failed to see distinct effect of
culture on the psychological variables, it was decided to design a further refined Study which was
done with the help of the feedback received from the participants in both countries.
78
4.5. Changes made after completion of Study 1
After completion of the survey in UK and Indonesia, all the feedback received were read and
every attempt was made to make sure that the feedback given by the participants was used to
develop a more refined version of the survey. All the feedback was exported from the Bristol
Online Survey (BOS) website into Microsoft Excel and was checked, after which amendments
were made. While the general feedback was that the survey was easy to follow, there were quite a
few comments on the instructions provided in the survey. For example, ‘Be more specific’,
‘Instructions may be clearer’, ‘the form confused me’ and ‘too many questions’. Informal chats
with participants after their participation also suggested that they required clearer instructions.
Although there was not a major change in the design of the survey however care was taken that all
feedback received was taken into consideration. In accordance with the feedback received the
instructions on the tasks were made clearer and were supported with example, e.g., for the
Facebook use scale where participants had to give an estimate of the time spent on Facebook per
day during the week and over the weekend in minutes, it was supported by example “1 hour = 60
minutes, so if you spend on average 1 hour on Facebook during the day, your response should be
60”. Additionally, there was some confusion on the difference between offline and online contacts.
Participants were given full description, e.g., “online social contacts are members of your social
network that you interact with online, i.e., on Facebook” and “offline social contacts are members
of your social contacts that you interact with face-to-face”. Questions where participants were
asked to enter the average time spent on FB, they were clearly informed that they should report
average Facebook use in Hours or minutes as required.
Additionally, while it was seen that the scale reliability of some of the items like the cultural
self-construal scale only achieved a medium scale reliability it was decided that all the twenty-four
items would now be included in the study as the reduced version might have affected the scale
reliability score. It was also decided that the individual self-construal items and the collectivism
measuring items would be kept as separate individual scales. Additional to this the items in the
self enhancement scale which measured the collectivist self attributes had a moderate scale
reliability (Cronbach alpha = .56), however for the items that measured the individualism
measuring attributes was high (Cronbach alpha = .70). It could be argued that the difference in the
values could be due to the small sample size, hence it was decided to be used in the final survey
79
as the items had been validated in past research (Gaertner et al., 2008). The attributes were further
validated in the pilot study conducted by Sedikides et al, 2003). After the amendments were
confirmed the final version of the survey was developed on BOS website and distributed in UK
and in Indonesia. The below section will now discuss the results and analysis of the final survey.
80
Chapter Five
Study 2: Survey (Facebook Responses)
5.1. Rationale for this study
To get valid and reliable responses from the UK and Indonesian sample, the amended version
of the survey was circulated with an aim to evaluate the impact of culture on various psychological
variables like online social support, online trust, online privacy, self enhancement and group
identity. Figure 5.1. illustrates the operational theoretical model which will be evaluated in the
below paragraphs based on the responses received from the participants in UK and Indonesia. The
model illustrates the various ways how the psychological variables (support, trust, privacy, group
identity and self-enhancement are affected by differences in cultural selves (individualism or
collectivism) that is mediated by the Facebook activities (Active: when members actively interact
with their social network on Facebook; Passive: when members interact passively with their social
network on Facebook) and Facebook frequency (FB Days: number of days members use
Facebook; FB HR: number of hours members use Facebook).
81
Figure 5.1.
Operational Theoretical Model showing all the variables and mediators
Note:
FB represents Facebook
FB Day, FB Active, Active and Passive are the mediators
FB Days represents number of days members used Facebook in the last 14 days
FB Time represents how many hours members have been on Facebook in the last twenty four hours
Active and Passive are the two different categories of Facebook interaction
82
To enhance understanding of the relationship between the variables another model (see Figure
5.2) below is illustrated that is used in all the four studies.
Figure 5.2.
Demonstrates the Basic Model that illustrates the relationship between Cultural Orientation,
Facebook Engagement and Psychological Outcome Variables
Note:
Facebook Engagement represents Facebook activities (Active and Passive), hours of Facebook use (FBHR) and days of Facebook use (FB Days)
Psychological Outcome Variables consists of Online Group Identity, Online Self-enhancement, Perceived Online Social Support, Online Trust and
Online Privacy Concern
5.2. Method
5.2.1. Participants
A total of 432 (UK: N = 209; 76.6 per cent female; Indonesia: N = 223) participants; more than
90 percent in Indonesia and more than 70 percent female in the UK participated in a survey and
were all Facebook users. The mean age of the participants was UK: 22.32 years (SD = 8.12) and
Indonesia: 19.34 years (SD = 1.16) with more than 80 per cent of the respondents falling in the
range from 18 to 21 years.
83
5.2.2. Procedure
Study 2, which was the refined version of Study 1 was developed online on the Bristol Online
Survey (BOS) website and was administered across university students in UK through the
University’s online system. The survey was also advertised on Facebook and Twitter. For the
Indonesian sample, an Indonesian research assistant was recruited who was fluent in both English
and their native language who helped to circulate the paper version of the survey to the University
students in Indonesia and then inputted the responses online on the BOS in English. Another
research assistant was recruited to monitor the data input process in order to avoid any mistakes.
While UK participants received the English version of the survey, the Indonesian participants
were given the translated version. The survey lasted for approximately 15-20 minutes. Full
participant consent was attained and participants were also made aware of their rights before their
participation in the survey. Participants were also given the chance to enter in a lottery draw to win
gift voucher as a thank you for their participation which was completely voluntary.
5.2.3. Materials & Item description
The survey consisted of several validated scales that was used to meet the research aim, which
will be discussed in the below paragraphs.
5.2.3.1. Cultural Self-Construal Scale: The items in this scale was a validated scale and was
adapted from Singelis (1994). Instead of the reduced version of the scale all the twenty-four items
in the scale was used for the study (12 items measuring collectivism self-construal and 12 items
measuring individualism self-construal). The internal consistency of the items was high for the sub
scale that measured the interdependent traits and had a Cronbach alpha of .80 (Facebook contacts)
and .76 (Overall social contacts). The internal consistency for independent traits measuring items
were moderate and had a Cronbach alpha of .62 (Facebook contacts) and .67 (Overall social
contacts), however as they were validated scales, it was decided to use in the study further.
5.2.3.2. Online Facebook Activity Scale: This was kept the same as Study 1. The scale used in
the survey was a validated scale adapted from Junco (2012), which had a high scale reliability
(Cronbach alpha = .89). The original scale adapted had 14 items in total which included items that
measured different aspects of Facebook activities. For this study the activities were divided into
84
two sub scales. Active interaction (which involved respondents to interact with other members on
Facebook actively) and Passive interaction (where members could use and monitor other members
and did not require the interaction with other members). Under the Active participation sub scale
– items included were a) Private messages, b) Commenting and c) Facebook chat which had a high
internal consistency (Cronbach alpha = .80). The Passive interaction sub-scale included items like
a) Playing games, b) Status Updates, c) Sharing Links, d) Checking up, e) Events, f) Posting
photos, g) Tagging photos, h) Viewing photos, i) Posting videos, j) Tagging videos and k) Viewing
Videos and also had a high internal consistency (Cronbach alpha = .85).
5.2.3.3. Online Group Identification Scale: The group identification scale looked at how much
respondents identified with their social group on Facebook was a validated scale and was adapted
from Brown et al. (1986) and was kept the same as was in the pilot study. The internal consistency
of the amended version of the six items in the group identification scale was high (Cronbach alpha
= .80) suggesting that the items measured the same construct.
5.2.3.4. Online Self-Enhancement Scale: This scale was designed as an attempt to show that self
enhancement does exist in both Western and Eastern culture. This in in support with past studies
by (Gaertner et al., 2008; Sedikides et al., 2003). The scales have been validated by Gaertner et al.
(2008) and for this study a similar method was employed to check the existence of self
enhancement in the online world. Respondents were asked to provide their truthful responses on a
six point Likert scale ranging from 0 (lowest score) to 5 (highest score). The higher the rating the
higher they score on that particular attribute. Participants were first asked to rate themselves on
the cultural traits, then their Facebook contacts followed by their rating for their Offline social
contacts (this could also include members that might also be on their Facebook contacts). In order
to find out if respondents rated themselves higher than their social contacts on Facebook, the rating
they provided for each category for themselves was subtracted from the ratings they provide for
their social contacts on Facebook. For example, if the respondent has rated on “Independent”
category as “5” for themselves and “3” for their social contacts on Facebook then “5-3 = 2” was
the difference that they had rated them self, i.e., they had rated themselves higher on that category.
In this way, all the individual category was calculated by using the Compute function in SPSS and
an overall rating for each category was calculated at the end. This was the rating that they gave
themselves in each category, which will help to check if they self-enhanced on each cultural
85
category based on their cultural orientation. It is to be noted that for this study analysis was only
conducted on the ratings respondents gave for themselves and their Facebook social contacts. The
internal consistency suggests a high Cronbach alpha of .71 for the collectivism measuring items
and moderate Cronbach alpha of .66 for the individualism measuring items (see Table 5.1 below).
5.2.3.5. Perceived Online Social Support Scale: This scale measured the social support and was
kept the same as was in Study 1. The items had been adapted from the validated MOS Social
Support Survey (Shelbourne & Stewart, 1991). Internal consistency of the items was very high
with a Cronbach alpha of .95 which confirms that the items measured the same construct.
5.2.3.6. Online Trust scale: This measured trust among members and was kept the same as Study
1. The items were adapted from the validated scale used by Krasnova et al. (2010). The internal
consistency of the items was high and had a Cronbach alpha of .88 (see Table 6 for further
information).
5.2.3.7. Online Privacy Concern Scale: This scale was developed to check the level of privacy
concern on Facebook among the respondents. This scale was developed after with discussion with
a focus group of three. The items were designed to fit in with the current study and would measure
the construct. The scale had seven items in total (six positively worded and two negatively worded
items) and was kept the same that was used in Study 1. The items were arranged so that respondents
have a mixture of positively worded and negatively worded items which will enforce some
cognitive understanding of the items and would avoid false responses. Respondents were asked to
give their truthful responses on the items which ranged from “Strongly Disagree” to “Strongly
Agree”. The items were coded using five point Likert scale ranging from “Strongly Disagree” to
“Strongly Agree”. For this study “Strongly Disagree” was coded as 1, “Disagree” was coded as 2,
“Neither agree or disagree” was coded as 3, “Agree” was coded as 4 and “Strongly Agree” was
coded as 5. The negative items were reversed scored using the compute into a different variable in
SPSS. The internal consistency of the items was low with a Cronbach alpha of .43. Both negative
items “I don’t care who looks on my Facebook profile” and “I get worried about people being able
to look on my personal profile on Facebook” were removed which increased the internal
consistency to Cronbach alpha of .70. Therefore, the final five items were retained for further
analysis (see Appendix 2 for a full list of items).
86
Table 5.1.
Reliability scores of the variables used in Study 2
Variable Cronbach alpha [α] Mean SD Number of
items
1) Facebook activity
a) Active .80 9.06 3.08 3
b) Passive .85 26.21 7.37 11
2) Social Support .95 30.62 9.70 10
3) Trust .88 18.69 4.13 6
4) Privacy .70 15.20 3.42 5
5) Group Identity .80 19.50 3.23 6
6) Self enhancement
a) Individualism categories .66 23.13 4.45 7
b) Collectivism categories .71 25.98 4.42 7
7) Cultural Orientation
a) Individualism .67 42.22 5.20 12
b) Collectivism .76 43.50 5.50 12
5.3. Results
This section will discuss the findings to understand the relationship between the predictors
(Individualism and Collectivism), mediators (active, passive, hours spent on Facebook in the last
24 hours (FB HR), number of days spent on Facebook in the last 14 days (FB Days)) and the
outcome variables (online group identity, online self-enhancement, perceived online social
87
support, online trust and online privacy concern). In short, the Operational model (Figure 5.1) is
explored individually and analysed (using the Basic Model – see Figure 5.2). This chapter is
divided into several sections. The first section will look at sample distribution through descriptive
statistics. The next section looks at the relationships of the variables through bivariate correlations
followed by a section on multivariate relationships where the various constructs are checked
whether they loaded on the same constructs using Maximum Likelihood Estimation (ML) using
SPSS version 22. The next section looks at the data which checks the mediation by using statistical
technique called Process and then the models are constructed and tested using Structural Equation
Modelling (SEM) using AMOS where the theoretical validity of relationships of the variables are
tested. SEM is used to test the relationship between the hypothesized model and the observed data.
Model fit is attained in AMOS which helps to support the findings. The final section is the
conclusion which outlines the limitations of the models and proposes plans for future analyses.
5.3.1. Descriptive statistics
Descriptive statistics is first used to test individual variable through frequency distribution and
explore options in SPSS which checks whether further in depth analysis should be carried out.
Frequency tables first explores the number of respondents in the data, their frequency counts. The
explore option looks at the missing data and outliers and looks at the kurtosis and skewness of the
data.
5.3.2. Data screening and preliminary analysis
Data was analysed using SPSS version 22. The threshold for significance was set to α = 0.05
for all decisions. Prior to in depth analysis data was first screened to check for missing data and
outliers. Boxplots and subsequent checks revealed the presence of some outliers and missing data.
All outliers and missing data were replaced with the mean for the variables. The assumptions of
normality were also checked if there was any presence of skewness or kurtosis in the data.
Normality of the data was accepted if the distribution of the data, i.e., both skewness and kurtosis
was within range (z < 2.58), (Hair et al., 2006).
88
5.3.3. Sample characteristics
Table 5.2 demonstrates the characteristics of the participants. The overall mean age of the UK
respondents was 22.32 (SD = 8.12) years and the age range were between 18 to 50 years. The
mean age of the Indonesian respondents was 19.34 (SD = 1.16) years. The age range for the
Indonesian respondents were 17 to 23 years. The age range for the UK respondents were more
varied than the Indonesian sample however overall sample consisted of more than 90 percent of
student population (full time and part time students included). This is also demonstrated in Table
5.3 which shows the variation of work for the sample in both UK and Indonesian respondents. The
highest number of respondents were from the population from 18 – 21 years of range which is also
reflected in both Table 5.2 and Table 5.3. The sample also showed a greater number of
participation form the female population in both UK and Indonesia.
Table 5.2.
Illustrates Descriptive statistics of participants in Study 2
Variable Mean (SD)/ Frequency Range/ Percentage
Overall (N= 432)
Age (years):
UK 22.32 (8.12) 18 – 50 (33)
Indonesia 19.34 (1.16) 17 -23 (6)
Country: UK: N = 209; Indonesia: N = 223)
Gender:
UK : Male 18 8.6
: Female 191 91.4
89
Indonesia : Male 66 29.6
: Female 157 70.4
Note: Overall represents both UK and Indonesia combined
Table 5.3.
Demonstrates Ethnicity and Work characteristics of participants in Study 2
Variable Frequency Percentage
Overall (N= 432)
Ethnicity:
UK: British 209
Indonesia: Indonesian 223
Work:
UK: (N = 209)
Employed (full time) 11 5.3
Employed (part time) 11 5.3
Unemployed 7 3.3
Student (part time) 11 5.3
Student (full time) 169 80.9
Indonesia: (N = 223)
Employed (full time) 1 .4
90
Employed (part time) - -
Unemployed - -
Student (part time) 3 1.3
Student (full time) 219 98.2
Table 5.4 below demonstrates the use of Facebook during the weekdays and weekend. The data
demonstrates that UK respondents use Facebook far more than the Indonesian respondents and
mostly use Facebook for social purpose (Weekday - Mean: 4.40; SD: 6.54) and (weekend – Mean:
4.89; SD: 14.13). Indonesian respondents also seem to use Facebook mostly for social purpose
(Weekday - Mean: .67; SD: 1.75) and (weekend – Mean: .57; SD: 2.17). There seems to be a
greater spread of Facebook use during the weekend for the UK sample (SD = 14.13). Similar time
on Facebook is invested during the weekend in both UK and Indonesia. There seems to be high
use of Facebook for work purpose among the Indonesian participants (Weekday - Mean: .44; SD:
3.28) and (weekend – Mean: .57; SD: 2.17) than the UK participants (Weekday - Mean: .23; SD:
6.54) and (weekend – Mean: .18 SD: .58).
Table 5.4.
Demonstrates Facebook use during weekday and weekends in UK and Indonesia
UK (N = 202) Indonesia (N = 212)
Study Work Social Mixed Study Work Social Mixed
Weekday:
Mean .76 .23 4.40 1.85 .48 .44 .67 1.02
SD 1.57 .65 6.54 2.69 .80 3.28 1.75 4.14
Weekend:
Mean .39 .18 4.89 1.79 .80 .57 1.03 1.27
91
SD .79 .58 14.13 2.58 2.01 2.17 1.85 2.19
Table 5.5 demonstrates the findings of the descriptive analysis of ‘the number of days and the
number of hours’ respondents have used Facebook in the last fourteen days. The results suggest
that UK respondents use at least 2 days in two weeks on Facebook unlike the Indonesian
respondents who seem to suggest that they might not use Facebook every day. T-test conducted
between the two countries revealed that there was a significant difference between UK (M = 12.96;
SD = 2.63) and Indonesia (M = 6.89; SD = 3.89), t=18.83, p < 0.001. The results further highlights
that the participants in the UK spend more time on Facebook than the Indonesian sample. There
was a significant difference in the number of hours’ participants used Facebook in UK (M = 4.31;
SD = 2.43) and Indonesia (M = 1.46; SD = 1.02), t = 16.01, p < 0.001).
Table 5.5.
Demonstrates Facebook use among UK and Indonesian respondents in Study 2
UK Indonesia Overall
Number of Days Facebook used in the last 14 days
Mean 12.87 6.89 9.78
SD 2.43 3.90 4.43
Minimum 2 0 0
Maximum 14 14 14
N 209 223 432
92
Average time (hours) spent on Facebook in the last
14 days
Mean 4.31 1.46 2.84
SD 2.43 1.02 1.53
Minimum 1 0 0
Maximum 13 8 13
N 209 223 432
Indonesian participants have more social contacts on Facebook (UK: M = 25.83; SD = 24.17;
Indonesia M = 47.45; SD = 72.42) and Offline (UK: 23.85; SD = .34; Indonesia: M = 38.82; SD
= 72.31) than UK participants but the Indonesian participants seemed to receive less social support
online (UK: M = 3.42; SD = .98; Indonesia: M =2.72; SD = .83). An independent sample t-test for
the Facebook contacts and social support score showed that the difference between sample was
significant both for number of social contacts (t = -4.08, df = 424, p < 0.05, two tailed and had a
medium effect size was (d = 0.40). The difference in the social support scores was also statistically
significant (t = 8.04, df = 430, p < 0.05, two tailed and had a high effect size (d = 0.77). This fails
to support H2 as higher number of Facebook contacts did not suggest higher social support online.
Indonesian participants have higher number of social contacts Offline too similar to online than
UK participants (UK: M = 23.85; SD = .34; Indonesia: M = 38.82; SD = 72.31) suggesting that
the Indonesian have a wider group of social contacts both online and offline. A further analysis
using an independent sample t-test between the mean total Facebook contacts and offline contacts
between the two countries confirms that the difference between the two groups was statistically
significant (Offline contacts: t = -2.70, df = 423, p < 0.05, two tailed and had a high effect size (d
= 0.51); Facebook contacts: t = -4.08, df = 424, p < 0.05, two tailed and had a medium effect size
(d = 0.40).
An independent sample t-test was run on the trust scores between UK and Indonesia which
confirmed that the UK participants scored higher on the trust scale (M = 3.14, SD = .69) than the
93
Indonesian sample (M = 3.09, SD = .69) as predicted however the difference in scores was not
statistically significant (t = .694, df = 430, p > 0.05, two tailed) thereby failed to support H5.
5.3.4. Investigating the relationship between variables
A Pearson’s bivariate correlation was conducted using SPSS version 22 to check the
relationship between the items for each construct and to check whether further analysis should be
carried out (see Table 5.6.).
Table 5.6.
Demonstrates inter-correlation of the variables in Study 2
SE CO Support Identity Trust Privacy Activity Gender Age
SE 1
CO -.197** 1
Support .033 .023 1
Identity -.082 .167** .092 1
Trust -.213** .140** .299** .296** 1
Privacy -.121* .126** .085 .187** .216** 1
Activity .198** -.246** .389** .023 .059 .026 1
Gender .127** -.148** .054 .045 -.035 .091 .123* 1
Age .050 .098* .067 .135** .149** .277** -.025 .069 1
Note: Correlation is significant at the 0.05 level (2-tailed) *
Correlation is significant at the 0.01 level (2-tailed) **
SE = Self enhancement, Support = social support and CO = cultural orientation
N = 432
Table 5.6. Suggests that there were significant relationships between variables used in the study.
For example, self enhancement seems to decrease with an increase in cultural orientation (r = -
94
.197, p < 0.01), i.e., increase in collectivism decreases self enhancement on Facebook. Increase in
self enhancement decreased feeling of online trust (r = -.213, p < 0.01) and privacy (r =.-.121, p <
0.05). However, increase in self enhancement increased activities on Facebook (r =.198, p < 0.01).
Increase in cultural orientation, i.e., the more collectivism members were the higher they
identified (r =.167, p < 0.01) and trust (r =.140, p < 0.01) the other members online. Increase in
cultural orientation increase privacy concern (r =.126, p < 0.01) on Facebook which is also quite
relevant to cultural traits as the more collectivist members were the higher they would be
concerned out their privacy. Increase in cultural orientation decreased Facebook activities (r = -
.246, p < 0.01), i.e., collectivists are active on Facebook then individualists.
Increase in social support significantly increased online trust (r =.299, p < 0.01) and Facebook
activities (r =.389, p < 0.01).
Increase in identification with their group members on Facebook seemed to increase their
feeling of trust (r =.296, p < 0.01) for their group members however it also increased their privacy
concern (r =.187, p < 0.01).
Increase in trust for online group members on Facebook increased in online perceived privacy
concern (r =.126, p < 0.01) suggesting that the level of trust for members might vary.
Age and gender were used as control variables as both had associations with the other variables.
For example, increase in gender significantly increased online self enhancement (r =.127, p < 0.01)
and Facebook activity (r =.123, p < 0.05) but significantly decreased cultural orientation (r = -.148,
p < 0.01).
Increase in age seems to significantly increased cultural orientation (r =.098, p < 0.01) online
group identification (r =.135, p < 0.01), online trust (r =.149, p < 0.01) and perceived online privacy
concern (r =.277, p < 0.01), which confirmed with age members can identify with their online
group members and due to the time and experience members have, they can develop trusting bonds
with their social network. However, higher age of members also made them more concerned about
their online privacy which could be since members might already have well established offline
social network and use the SNSs just for causal socialising.
95
5.3.5. Check for Moderation
Once the relationship of all the variables in the study was established, attempt was made to
check if there was any moderation effect. As the data was collected in two different cultures, it
was necessary to check for any moderating effect by culture on the variables. PROCESS version
2.15 is an Add-ins in SPSS was used for the analysis. It uses ordinary least squares or logistic
regression based framework to estimate direct and indirect effects for single and multiple
mediators and moderators with the help of several models, it uses Bootstrap and Monte Carlo
confidence intervals, which are also provided along with several measures of effect size which
help to make more accurate inferences. It also provides the option to check Sobel test result in the
same output which helps to check the result with the traditional method. These benefits outperform
the traditional method of checking such effects by using the traditional method in SPSS using
logistic regression as it does not allow check such effect using several variables.
When the relationship between two variables depends on the presence of any third variable, a
moderation analysis (also known as interaction) is required. This is certainly a very important
aspect of this study as the data was collected in two different countries, namely UK and Indonesia.
As such it was important to check whether country played any moderating effect between the
predictor and outcome variables. The relationship between two variables is said to be moderated
when a third variable (in this case country) effects its size and sign. A three-way mediated
moderation analysis was run in PROCESS with the direct path ran from predictor (X) to mediator
(M) and then another path ran from M to Outcome variable (Y) and a direct path from X to Y. All
the three paths were also moderated by country such that the effect between each path depends on
the effect from the moderator and for this study the moderator was “Country”. Each variable
namely “Facebook activity (both active and passive”, “Social support”, “Trust”, “Privacy”,
“Identity” and “Self enhancement”. A moderation analysis ran using the PROCESS plug in SPSS
confirmed that country did not have a strong moderating influence on majority of the outcome
variables. Out of the six outcome variables there was only one significant moderating effect by
country was established (see Table 5.7. below) due to which further moderation analysis was
discarded. Instead, mediation effect of engagement on SNS on the psychological variables used in
the thesis. The below section will discuss the findings of the mediation analysis.
96
5.3.6. Mediation Analysis
As no moderation effect by country was established, simpler models using mediation analysis
using PROCESS, a widely used method used in psychological researches (Baron & Kenny, 1986;
Hayes, 2009; Hayes, 2012). The next aim was to establish the causal relationships of other
intervening variables that might affect the relationship of the predictor variables and the
psychological variables. These intervening variables are called mediating variables. This variation
in the predictor variable (X) causing a variation in the mediating variable or variables (M) which
in turn will causes a variation in the outcome variables (Y) could be analysed. To test this effect
PROCESS estimates the path from X to M (direct) and M to Y (indirect).
The cultural self-construal of the participants which has been termed as ‘Cultural Orientation’
(CO) was used as a predictor in the analysis and each outcome variables (perceived online social
support, online group identity, online self-enhancement, online self-enhancement, online trust and
online privacy concern) were tested without the mediator first and then with the mediators. At first
the direct relationship between the predictor variable or IV and the outcome or the DV was checked
and then the mediators were introduced in the analysis. After running the single mediator analysis,
a combined mediator analysis was also run, where all the four mediators (FB Active, FB Passive,
FB Days and FB HR) were analysed at the same time. Unlike SPSS which uses the traditional
regression analysis that allows to check only one mediator at a time, PROCESS gives the
advantage of checking more than one mediator at once. Age and gender were used as control
variables. (see Table 5.7. for the full results of the mediation analysis for each construct).
Table 5.7.
Demonstrates the Indirect Effect of Cultural Orientation on Outcome Variables
Variable b SE Lower CI Upper CI
Perceived Online Social Support (R2 = .244)
Active -.142 .048 -.259 -.063
Passive -.063 .038 -.149 .001
97
FB Days -.101 .044 -.200 -.026
FB HR -.038 .035 -.120 .022
Online Trust (R2 = .073)
Active -.050 .033 -.125 .007
Passive -.017 .028 -.074 .036
FB Days .021 .031 -.036 .086
FB HR .004 .022 -.039 .048
Online Privacy (R2 = .144)
Active -.006 .028 -.062 .047
Passive -.058 .027 -.118 -.013
FB Days .057 .028 .009 .121
FB HR .046 .026 .000 .103
Online Group Identity (R2 = .057)
Active -.002 .025 -.053 .044
Passive -.013 .022 -.059 .030
FB Days -.012 .025 -.065 .034
FB HR .000 .021 -.042 .041
Online self-enhancement
Collectivist attributes (R2 = .094)
Active -.062 .030 -.135 -.012
Passive .016 .025 -.034 .068
FB Days -.070 .033 -.148 -.013
FB HR .004 .028 -.056 .057
98
Individualist attributes (R2 = .013)
Active .020 .030 -.034 .086
Passive -.019 .027 -.077 .034
FB Days .000 .028 -.057 .055
FB HR -.011 .026 -.067 .036
Table 5.7 above is a summary of the findings of the results of the combined mediation analysis
for all the outcome variables using PROCESS. The results confirm that there was a significant
mediation effect for social support F(7,424) = 19.57, p < 0.05, R2 = .24, online trust F(7,424) =
4.75, p < 0.05, R2 = .073, online privacy concern F(7,424) = 10.23, p < 0.05, R2 = .144, online
group identity F(7,424) = 3.65, p < 0.05, R2 = 0.57 and for online self-enhancement on collectivist
attributes F(7,424) = 6.30, p < 0.05, R2 = .094. However the mediation analysis for the self-
enhancement on individualist attributes was non-significant F(7,424) = .83, p > 0.05, R2 = .013.
Although the details for the single mediator mediation analysis has not been provided, the results
for the combined models shows indirect effect of cultural orientation on the outcome variables.
Therefore, it was further decided that such mediation analysis be explored further which will help
to show in-depth results and help to develop models to understand how the various relationships
between the variables work. The below sections will be discussing the process and the analysis of
mediation analysis using AMOS.
5.3.7. Modelling of Multivariate Relationships
The below sections will be an attempt to outline the various ways how the various multivariate
relationships were evaluated through various statistical models.
5.3.7.1. Preparatory analyses
Exploratory Factor Analysis help to understand the underlying variables from a large set of data
and further help to understand the relationships among the variables. Five outcome variables are
99
proposed for categorization, namely, trust, social support, group identity, privacy and self
enhancement along with the predictor variable (cultural orientation) which was further subdivided
divided into collectivism and individualism. There were four mediating variables, namely,
Facebook activity (divided as active and passive), FB Days (which looked at the number of days’
members had been on Facebook within the last 14 days) and FB hours (which looked at the number
of hours’ members had been on Facebook in the last 14 days). An exploratory factor analysis was
conducted to establish the theoretical importance these variables could establish. The factor
analysis was implemented using the Maximum Likelihood (ML) procedure using the Promax
method.
The descriptive statistics and correlations were conducted to alleviate any assumption failure
of univariate and multivariate distributions and the results confirmed that all the items were
normally distributed. Sample adequacy was measured using the Kaiser-Meyer-Oklin (KMO)
technique. A Bartlett’s Test of sphericity was used to a sufficient relationship between the items
confirming further analysis could be conducted.
Promax method was used which extracted items with values greater than 1.0 and in a one factor
loading was mostly preferred so that the variables could be used for structural equation model.
The exploratory factor analysis would help to establish how the indicators in a latent variable helps
to explain the theoretical constructs with the help of structural equation modelling. This was
achieved using analysis of moment structures (AMOS). Data reduction was conducted in factor
analysis to make sure that the items measured the factor (see Table 8 to check the amended version
of the variables). Items that cross-loaded on another factor or did not load on any factor were
deleted from the analysis. Once the latent variables were established their reliability scores were
also measured.
5.3.7.2. Reliability Measure
Internal consistency was checked using the scale reliability option in SPSS which looks at the
Cronbach alpha [α] to determine that the scales used are reliable, i.e., they are measuring what they
are supposed to. A series of reliability checks were conducted for each latent variable (see Table
5.8.).
100
Table 5.8.
Demonstrates the original items and the amended number of items (N) for each variable, their
Cronbach alpha (α), the Mean and SD of the amended variables
Variable Original items N
(Cronbach alpha
[α])
Amended Items N
(Cronbach alpha
[α])
New Mean New SD
Facebook activity
Active 3 (α = .80) 3 (α = .80) 9.06 3.08
Passive 11 (α = .85) 5 (α = .76) 14.41 4.09
Social Support 10 (α = .95) 6 (α = .91) 18.63 5.64
Trust 6 (α = .88) 5 (α = .85) 15.65 3.44
Privacy 5 (α = .70) 3 (α = .73) 9.31 2.50
Group Identity 6 (α = .80) 4 (α = .79) 13.44 2.14
Self enhancement
Individualism categories 7 (α = .66) 4 (α = .71) 14.31 3.22
Collectivism categories 7 (α = .7ive 1) 5 (α = .74) 19.67 3.48
Cultural Orientation
Individualism 12 (α = .67) 6 (α = .61) 21.51 3.22
Collectivism 12 (α = .76) 6 (α = .66) 23.28 3.00
Table 5.8. above demonstrates that all the variables used in the analyses to as subsequent latent
constructs or as observed indices of the theoretical constructs has acceptable internal consistency
ranging from (α = .61 to α = .91), Blacker & Endicott (2002) even after the items in the latent
101
constructs were amended. Apart from the privacy scale, all the other scales had been validated in
past studies. However, the privacy scale used had acceptable reliability score (see Table 5.8), hence
was used in the study. The below sections will discuss the amended variables used for the models.
5.3.8. Online Facebook Activity Scale: This validated scale had 14 items in total. While Pearson
bivariate correlation confirmed that most of the items correlated with each other. However
Exploratory Factor analysis confirmed that the items loaded on three different factors. On close
look at the individual items it was clear that there was not a very clear differentiation between each
loading which made it difficult to confirm what factors they were. For example, “Viewing videos”
loaded on Factor one and other two video items “Tagging videos” and “Posting videos” also loaded
on another factor. Similarly, categories related to photos also loaded on two factors. Due to the
difficulty in coming up with individual factors that measured a construct, it was decided to divide
Facebook activity to two apparent categories “Active interaction– where members interacted with
other members and the presence of other member is needed” “Passive interaction – in this type of
interaction members don’t need the presence of other members”. For this study, further on from
now Facebook activity will be divided into two different categories “Active” and “Passive”. For
this study inter item correlation with < .50 and above might affect model fit in Structural Equation
Modelling (SEM) and hence decided to be removed from the scale as it would suggest that both
items meant the same or were very similar.
The “Active” scale consisted of items like “Private messages”, “Commenting” and “FB chat”.
Inter-items correlation confirmed that the items did not load very highly however had a correlation
of .4 had and above and hence were included in the scale suggesting high factorability. Exploratory
Factor Analysis using the Maximum Likelihood method (ML) suggested a Kaiser-Meyer Olkin
measure for sampling adequacy was .64 and Bartlett’s test of sphericity was significant p < .05
however Extracted Factor Rotation could not be achieved as only one factor loading was achieved.
Finally, a Promax extraction method confirmed one factor loading and eigan values of 1 indicated
one factor loading with a cumulative variance of 59.59 %. Factor loadings of the items were .5 and
above. Internal reliability of the items was checked using Cronbach alpha (a) which had a high
internal consistency of a = .80.
The items in the “Passive” scale originally consisted of the remaining 11 items, however
bivariate correlation suggested that some of the items had high multi-collinearity and hence were
102
eliminated from the scale. It is however important to note that most of the items correlated < .3
suggesting that there is good factorability. This was done so with an aim to propose a model in
SEM with accepted model fit. The remaining items five items were “Sharing links”, “Checking
up”, “Events”, “Posting Photos” and “Viewing Photos”. Factor Analysis using the Maximum
Likelihood method (ML) suggested a Kaiser-Meyer Olkin measure for sampling adequacy was .78
and Bartlett’s test of sphericity was significant χ2 (5) = 23.13, p < .001). Finally, a Promax
extraction method confirmed one factor loading and Eigen values of 1 indicated one factor loading
with a cumulative variance of 40.87 %. Factor loadings of the items were .4 and above. Internal
reliability of the items was checked using Cronbach alpha (a) which had a high internal consistency
of a = .76, (Blacker & Endicott, 2002).
5.3.9. Online Cultural Self-Construal Scale: The original scale had 12 items under measuring
individualism traits and 12 items measuring collectivism traits. A Pearson Bivariate correlation
confirmed that items “all items under this sub scale did not have high multi-collinearity of .5 and
above and hence was not overly concerned however most of the items correlated > .3 suggesting
that there was good factorability. Exploratory Factor Analysis was conducted at first on the
collectivism measuring traits which suggested that item “My happiness depends on the happiness
of those around me”, “Even when I strongly disagree with my social group members, I avoid an
argument” and “I respect people who are humble about themselves” had low community of < .3
and hence were eliminated. EFA was conducted on the final nine items (see Appendix for a full
list) using the Maximum Likelihood method (ML). A Kaiser-Meyer Olkin measure for sampling
adequacy was .78 and Bartlett’s test of sphericity was significant at χ2 (12) = 29.92, p = .003).
Finally, a Promax extraction method confirmed three factor loading and Eigen value of 1 indicated
three factor loading with a cumulative variance of 26.52%, 33.45 % and 38.50 %. Factor loadings
of the items were .3 and above. Internal reliability of the items was checked using Cronbach alpha
(a) which had a high internal consistency of a = .74.
Similarly, the individualism measuring items were now checked using Pearson Bivariate
correlation which revealed that item “I feel comfortable using someone’s first name soon after I
meet them, even when they are much older than I am” correlated negatively on some of the other
items hence was eliminated. Also, items “I am the same person at home that I am during social
gathering” and “I act the same way no matter who I am with” had high multi-collinearity = .50 and
103
hence item “I am the same person at home that I am during social gathering” was eliminated from
further analysis. EFA was conducted on the final nine items (see Appendix 2 for a full list) using
the Maximum Likelihood method (ML). A Kaiser-Meyer Olkin measure for sampling adequacy
was .78 and Bartlett’s test of sphericity was significant at (χ2 (12) = 29.92, p = .003). Finally, a
Promax extraction method confirmed three factor loading and Eigen value of 1 indicated three
factor loading with a cumulative variance of 26.52%, 33.45 % and 38.50 %. Factor loadings of
the items were .3 and above. Internal reliability of the items was checked using Cronbach alpha
(a) which had a high internal consistency of a = .74.
5.3.10. Online Group identity: The original scale had 7 items measuring the construct. A Pearson
Bivariate correlation confirmed that items “I see myself as a member of my Facebook community”
and “It is important for me to be a member of my Facebook community” had high multi-
collinearity (r = .68) and hence item “It is important for me to be a member of my Facebook
community” was eliminated from the scale however most of the items correlated < .3 suggesting
that there was good factorability. Exploratory Factor Analysis was conducted on the remaining six
items using the Maximum Likelihood method (ML) which suggested that “I see myself as a
member of my Facebook community” had a low community of .197 and hence was eliminated and
the EFA was run again with the final four items “I like being a member of my Facebook
community”, “I am not glad to be a member of my Facebook community” (reversed scored), “I
am not proud to be a member of my Facebook community” (reversed scored) and “I do not like
being a member of my Facebook community (reversed scored). A Kaiser-Meyer Olkin measure
for sampling adequacy was .75 above and Bartlett’s test of sphericity was significant at (χ2 (2) =
8.72, p = .013). Finally, a Promax extraction method confirmed one factor loading and Eigen value
of 1 indicated one factor loading with a cumulative variance of 50.22 %. Factor loadings of the
items were .6 and above. Internal reliability of the items was checked using Cronbach alpha (a)
which had a high internal consistency of a = .79.
5.3.11. Online Self enhancement: The original scale had 7 items under measuring individualism
traits and 7 items measuring collectivism traits. A Pearson Bivariate correlation confirmed all
items under both sub scale did not have high multi-collinearity of .5 and above and hence was not
overly concerned however most of the items correlated > .3 suggesting that there was good
factorability. Exploratory Factor Analysis was conducted at first on the individualism measuring
104
traits which suggested that item “Separate” had a low community of <.3 and hence was eliminated.
Further analysis also suggested that item “Unconstrained” did not load on any factor and hence
was also eliminated. EFA was now run on the final four items “Independent”, “Leader”, “Unique”
and “Original” using the Maximum Likelihood method (ML). A Kaiser-Meyer Olkin measure for
sampling adequacy was .74 and Bartlett’s test of sphericity was significant at (χ2 (2) = 6.77, p =
.034). Finally, a Promax extraction method confirmed one factor loading and Eigen value of 1
indicated one factor loading with a cumulative variance of 39.53 %. Factor loadings of the items
were .6 and above. Internal reliability of the items was checked using Cronbach alpha (a) which
had a high internal consistency of a = .71.
Similarly, Exploratory Factor Analysis was conducted on the 7 collectivism measuring traits
which suggested that item “Self-sacrificing”, “Compromising” and “Loyal” had a low community
of <.3 and hence was eliminated. EFA was now run on the final four items “Respectful”,
“Compliant”, “Tolerant” and “Modest” using the Maximum Likelihood method (ML). A Kaiser-
Meyer Olkin measure for sampling adequacy was .76 above and Bartlett’s test of sphericity was
non-significant at (χ2 (2) = .703, p < .05) however chi square (χ2) is influenced by sample size and
hence its significance is highly debatable. Finally, a Promax extraction method confirmed one
factor loading and Eigen value of 1 indicated one factor loading with a cumulative variance of
40.38 %. Factor loadings of the items were .5 and above. Internal reliability of the items was
checked using Cronbach alpha (a) which had a high internal consistency of a = .72.
5.3.12. Perceived Online Social support: The original scale had 10 items measuring the
construct. A Pear Bivariate correlation confirmed that some of the items had high multi-
collinearity (r = .7 and above) and hence were eliminated from the scale however most of the items
correlated > .3 suggesting that there was good factorability. The remaining items six items were
“Someone you can count on to listen to you when you need to talk”, “Someone who provides you
with information to help you understand a situation”, “Someone to give you good advice on crisis”,
“Someone whose advice you really want”, “Someone to share your most private worries and fears
with” and “Someone with whom you can have a good time with”. Factor Analysis using the
Maximum Likelihood method (ML) suggested a Kaiser-Meyer Olkin measure for sampling
adequacy was .91 above and Bartlett’s test of sphericity was significant at (χ2 (9) = 40.88, p <
.001). Finally, a Promax extraction method confirmed one factor loading and Eigan values of 1
105
indicated one factor loading with a cumulative variance of 64.64 %. Factor loadings of the items
were .6 and above. Internal reliability of the items was checked using Cronbach alpha (a) which
had a high internal consistency of a = .91.
5.3.13. Online Trust: The original scale had 6 items measuring the construct. A Pear Bivariate
correlation confirmed that one of the item had high multi-collinearity (r = .7) and hence was
eliminated from the scale however most of the items correlated < .3 suggesting that there was good
factorability. The remaining five items were “Do their best to help me”, “Do care about the
wellbeing of others”, “Are open and receptive to the needs of each other”, “Are honest in dealing
with each other”, “Keep their promises” and “Are trustworthy”. Factor Analysis using the
Maximum Likelihood method (ML) suggested a Kaiser-Meyer Olkin measure for sampling
adequacy was .85 above and Bartlett’s test of sphericity was significant at (χ2 (9) = 152.62, p <
.001). Finally, a Promax extraction method confirmed one factor loading and Eigen value of 1
indicated one factor loading with a cumulative variance of 55.25 %. Factor loadings of the items
were .6 and above. Internal reliability of the items was checked using Cronbach alpha (a) which
had a high internal consistency of a = .88.
5.3.14. Online Privacy Concern: The original scale had 7 items measuring the construct. A
Pearson Bivariate correlation confirmed that two of the items “I don’t care who looks on my
Facebook profile” and “I get worried about people being able to view my personal information on
Facebook” which were reversed scored correlated negatively hence were eliminated from the
scale. Items “I like to use filter settings to group my social contacts on Facebook” and “Using filter
settings is important for me as it helps me to be open in my opinions on Facebook” had a high
multi-collinearity of .60 and hence item “Using filter settings is important for me as it helps me to
be open in my opinions on Facebook” was eliminated which confirmed that item “. An Exploratory
Factor Analysis was conducted using the Maximum Likelihood (ML) method which confirmed
that item “I like to use filter settings to group my social contacts on Facebook” did not load on any
factor and hence was eliminated from the analysis. The remaining items three items were “I keep
myself up to date with changes in privacy settings”, “I always update my security settings on my
Facebook account” and “I am fully aware of the use of privacy settings on Facebook”. A Kaiser-
Meyer Olkin measure for sampling adequacy was .64 above and Bartlett’s test of sphericity was
significant (p < .001) however as only one factor was extracted, factor extraction could not be
106
achieved. Finally, a Promax extraction method confirmed one factor loading and Eigen value of 1
indicated one factor loading with a cumulative variance of 51.91 %. Factor loadings of the items
were .5 and above. Internal reliability of the items was checked using Cronbach alpha (a) which
had a high internal consistency of a = .74. Table 5.9 illustrates the full list of factor loadings of
variables in Study 2. The below section 5.1.9 will outline the SEM analysis.
Table 5.9.
Demonstrates variables with their factor loadings, percentage of variance, number of items (N)
and their Cronbach alpha score [a]
Variable Rage of Factor loadings Percentage of variance Number of items
(Cronbach alpha [a])
Facebook activity
Active .55 to .90 59.59 3 (α = .80)
Passive .42 to .80 40.87 5 (α = .76)
Social Support .68 to .89 64.64 6 (α = .91)
Trust .60 to .88 55.25 5 (α = .85)
Privacy .53 to .89 51.91 3 (α = .73)
Group Identity .55 to .87 50.22 4 (α = .79)
Self enhancement
Individualism
categories
.50 to .72 39.53 4 (α = .71)
Collectivism
categories
.57 to .71 40.38 5 (α = .74)
107
Cultural Orientation
Individualism .30 to 1.00 13.16/ 29.89/ 34.73 6 (α = .61)
Collectivism .30 to .65 26.54/ 33.45/ 38.49 6 (α = .66)
5.3.15. SEM Analysis
Structural Equational Modelling (SEM) is used to measure relationships between constructs
variables depicted in the various models (Arbuckle, 2006). SEM was chosen as the method of
analysis as it has the benefit of being able to test the relationship between severable constructs at
the same time and their observed measures. It also helps to control for measurement errors to
establish relationships that are proposed by the model. The iterations help to assess models that bit
fits the data (Schumacker & Lomax, 2004a). Modelling using SEM was conducted using five step
approach, namely, a) model specification, b) model identification, c) model estimation, d) model
testing and e) model modification which is briefly described for each model.
5.3.16. Model Specification
The model specification, its design and measurement has been based on theoretical evidence
(Byrne, 2001) and items used in the measurement which is decided had been considered after
conducting literature review on the topic, where the items had been used in past research. This
ensures that the model has good construct validity. As mentioned in the earlier section that apart
from the privacy scale all the scales had been adapted from past validated scales that had been
used in past study. Each construct is used as a latent variable as such construct were measured
through items that were observed, in this study it was through self-report responses through an
online survey.
The model consists of six constructs that were measured through observed items. They were
cultural orientation of respondents was defined by two latent constructs namely individualism and
collectivism and were the exogenous variables. Online self-support, online trust, online privacy,
108
group identification and online self enhancement were the endogenous variables. Facebook
activity was defined by two observed variables ‘Active and Passive’, Facebook time measured the
average time respondents spent on Facebook in the last 14 days, which were used as mediators in
the model. In order fully understand Facebook engagement, it was important to also understand
the duration of Facebook use among its members. Therefore, it was decided to include two
additional observed variables (FB Days (measured the average number of days members used
Facebook in the last fourteen days) and Facebook HR (measured the average number of hours
members used Facebook in the last fourteen days). It is important to point out that as not all
members indulged in use of Facebook every day and not every member uses Facebook for the
same length of time, it was important to use the observed variables separately, which would help
to achieve a more in-depth understanding of Facebook engagement. The error term in the model
identifies the amount of variance in each observed variable
5.3.17. Model Estimation
Model estimation was a process which helped to check for relationships between items and or
constructs. AMOS was used to obtain Model Fit indexes. The indexes that we used to access the
model fit in this study were chi square (X2), Comparative Fit Indices (CFI), Normal Fit Index
(NFI) and the Root Mean Square Error of Approximation (RMSEA). Model fit for CFA and SEM
was checked across a range of fit indices (Hooper, Coughlan, & Mullen, 2008).
5.3.17.1. The chi square (X2): Provides the observed and implied model difference. The aim of
the model was to attain a non-significant effect (Chi square - X2 = p > 0.05) as a non-significant
chi-square would suggest that the implied theoretical model matches the original sample variance
matrix however as X2 is sensitive to sample size it tends to provide statistically significant result
based on the sample size. It is for this reason that other fit indexes were also used to check for
model fit.
5.3.17.2. Goodness-of-Fit Index (GFI): The GFI measured the amount of variance and
covariance in the sample matrix that is produced by the proposed matrix i.e., it is the ratio of the
sum of the squared differences between the observed and the reproduced matrices to the observed
109
variance. This index is similar to the R2 in a regression output. An acceptable range for GFI is .95
and above (Schumacker & Lomax, 2004).
5.3.17.3. Root Mean Square Residual Index (RMR): The RMR index is used to compare two
different models with same dataset. It uses the square root of the mean squared differences between
the original sample matrix and the proposed matrix. The acceptable range for RMR is anything <
0.05 (Jöreskog & Sörbom, 1982).
5.3.17.4. Normal Fit Index (NFI) and Comparative Fit Index (CFI): NFI and CFI rescales the
chi-square to a fit range from 0 (no fit) to 1 (perfect fit) range (Bentler, 1990) suggesting that the
closer the index is to 1 the better is the model fit.
5.3.17.5. Root Mean Square Error of Approximation (RAMSEA): This model fit Index is
based on non-centrality measure and also rescales the chi-square complexities and model fit ranges
from 0.01 (excellent), 0.05 (good) and 0.08 (medium) indication of model fit (MacCallum,
Browne, & Sugawara, 1996; Steiger, 1998)
5.3.18. Model Testing
Once the model has been run, its fit indexes were checked and if the model fit indexes didn’t
fall under the acceptable range, the model was re-specified by adding or deleting indicators or
variables. After making the required adjustments each model was checked again and this process
of re-specification and checking was done till each model reached the final model fit.
5.3.19. Model Modification
The final step in the model re-specification is by modification of the model by adding or
deleting paths. This was achieved by constantly checking the regression weights of the paths and
checking the modification suggestions.
5.3.20. Re-specification
The models in the study were specified to enhance the FIT indexes however care was taken that
any changes or alterations made to the model was theoretically supported (Meyers, Gamst, &
Guarino, 2006). The below method was used to re-specify a model and the same process was
followed for all the models in the study.
110
a) Determine all the constructs that should be included in the model. This was achieved by previous
literature review, past findings and also supported current research hypotheses.
b) At first single models were drawn in AMOS.
c) Results were checked along with the FIT indices.
d) Checked regression weights and significant and non-significant paths.
e) Checked modification suggestions.
f) Modified models based on modification suggestions.
g) Ran all models again.
h) Re-examined the fit measures.
i) Made decision whether to keep the model or make further modification depending on the results
outcome (Meyers et al., 2006; Schumacker & Lomax, 2004). The final models are briefly
discussed below.
5.3.21. Process
Step 1:
At first a construct was selected and a single model which consisted of a predictor variable or
IV, an outcome variable or DV and a mediator were designed using AMOS. While this research
wanted to access the effect of Facebook use on the psychological outcome variables, it was decided
to break Facebook use in days and time. It is believed that this will help to capture the effect of
Facebook in more detail. Therefore, the design consisted of four mediators, namely FB Active
(members have actively interacted with other members), FB Passive (members indulge in
browsing through the site without actually interacting with anyone), FB Days (measures the
number of days respondents had been on Facebook in the last fourteen days) and FB HR (measures
the number of hours’ respondents had been on Facebook in the last fourteen days), which was used
in a single model, one at a time. This allowed to check for the effect of each mediator on the
relationship of the IV and DV. The IV were the two dimensions of cultural self-construal
111
(collectivism/ individualism) which formed the cultural orientation (CO) of the participants. Each
CO was analysed separately using each of the four mediators separately.
Step 2:
Once the single models were run, the four mediators were then combined into a combined
model. Each IV (collectivism and individualism) was analysed using the mediation analysis
function in AMOS and their model fit was evaluated.
Step 3:
Attempt was also made to combine both CO (individualism and collectivism) together so that
there was one IV with a hope to simplify the design. However, on running AMOS it was
established that the model fit failed to meet the required acceptable fit indices. Therefore, it was
decided that each construct collectivism and individualism (which together make up one’s cultural
self-construal) will be measured separately. Due to the several number of models in the study, all
the single models had been discussed below and a brief outline of the analysis of the combined
model has been presented. While the single models can be found for each construct, the combined
models are presented in the Appendices. Furthermore, attempt has been made to include several
model fit indices to enhance model fit acceptance. Mediation analysis was conducted across all the
models with 2000 bootstrapped samples (Shrout & Bolger, 2002) with 95 % confidence interval.
The results for each construct will be discussed in the following order:
a) Online Group Identity
b) Online Self-Enhancement
c) Perceived Online Social Support
d) Online Trust
e) Online Privacy Concern
5.3.21.1. Online Group Identity
This model looked at the impact of cultural orientation (individualism/ collectivism) on how
much members identified with their group members on Facebook. While Models 1,2,3,4 (Figure
112
5.3) and Model 5,6,7,8 Figure (5.4) are the single mediator models and Figure 5.5 (see Appendix
7) combined models where all the four mediators are analysed together for each cultural
orientation.
H1: Facebook activities (type) and frequency (time) will mediate the relationship between
culture and group identity such that increase in Facebook activity will increase the level of
group identity.
The number of days spent on Facebook (Model 3: FB Days) and the number of hours’ members
spent on Facebook (Model 4: FB HR) had a significant mediating effect on group identification
and confirms full mediation (see Table 5.10) as there also seems to be no direct effect from
collectivism to group identity (Model 3: beta = .06, p > .05; Model 4: beta .03, p > .05). Increase
in collectivism significantly decreased FB activities in all the models. Increase in FB Days, i.e.,
the number of days’ members spend on Facebook and increase in FBHR, i.e., the number of hours’
members spend on Facebook also increased group identity. This suggests that time spent on
Facebook seems to help in group identification for members with collectivist attributes and not
what type of activity they do which is also supported by acceptable Model fit indexes (see Table
5.10) partially supporting H1.
113
Figure 5.3.
Models 1 to 4 Illustrates Standardized Path Coefficients of the Mediating Effect on Online Group
Identity using Collectivism as the predictor variable
Note: FB = Facebook; FB Active = active interaction on Facebook; FB Passive = passive interaction on Facebook; FB Days = average number of days’
respondents had been on Facebook in the last 14 days and FB HR = average number of time (in minutes) respondents had been on Facebook in
the last 14 days
Figure 5.4 which uses individualism as the cultural orientation as the predictor variable
demonstrates that in general increase in individualism decreases activities on Facebook. FB Active
seems to have a non-significant effect on group identity (beta = .01, p > 0.05) suggesting there is
no effect of active participation on Facebook on group identity. Such effect seems to be consistent
with both cultural orientations (collectivism and individualism). It is possible that such activities
114
require more one-to-one interaction where the entire group is not involved. There also seems to be
a very weak direct effect from FB Passive to group identity (p > .05) and the confidence intervals
was (lower: -.059 and upper: .149) confirming that although the indirect effect was just acceptable
(beta = .053), there was no mediation. Model 7 however confirms that the direct path from FB
Days (which measures the number of days one spends on Facebook) and (Model 8) FB HR (the
number of hours one spends on Facebook) was now significant suggesting the number of days’
members spent and the time spent on Facebook seems to effect group identity however the effect
seems to be weak in both the models (Model 7: b = .005, p > 0.05, -.075, -.016 and Model 8: b =
.017, p < 0.05, - 0.45, - .006. (See Table 5.11) also partially supports H1.
Figure 5.4.
Models 5 to 8 Illustrates Standardized Path Coefficients of the Mediating Effect on Online
Group Identity using Individualism as the predictor variable
115
Note: FB = Facebook; FB Active = active interaction on Facebook; FB Passive = passive interaction on Facebook; FB Days = average number of days’
respondents had been on Facebook in the last 14 days and FB HR = average number of time (in minutes) respondents had been on Facebook in
the last 14 days
Figure 5.5. (see Appendix 7) demonstrates the combined model which consists of all the four
mediators (FB Active, FB Passive, FB Days and FBHR) and also tests both cultural orientation
(collectivism and individualism) separately. The results suggest that in both Model 9 and Model
10, the direct link between collectivism and group identity (beta = .80, p > .05) and individualism
and group identity (beta = 1.94, p > .05) is non-significant suggesting that both cultural orientation
does not help to predict group identity. In Model 9 the increase in collectivism significantly
decreased activities on Facebook (b range is -.72 to -.87) and in Model 10 (b range is -.70 to -.84)
which is quite strong. However, all the direct paths from Facebook activities in both the models
seems to be non-significant (collectivism: FB Active – beta -.14, p > .05, FB Passive – beta .32, p
> .05, FB Days – beta .02, p > .05 and FB HR – beta .03, p > .05); Individualism: FB Active – beta
-.15, p > .05, FB Passive – beta .42, p > .05, FB Days – beta .02, p > .05 and FB HR – beta .03, p
> .05). Both the combined models Model 9 (b = .263, p > 0.05, -1.496, .235) and Model 10 (b =
.204, p > 0.05, -2.653, .211) failed to show any significant mediating effect. Table 5.10. below
also confirm that the Model fit scores also go beyond the acceptable range for all the fit indexes
when the mediators are combined together in one model. There was no mediating effect established
in both the model which also reflects in the confidence intervals.
Table 5.10.
Illustrates the Model Fit Indices, the Mediation Effect (Indirect Effect) & the Confidence
Intervals for all the models for Online Group Identity
Model RMR GFI NFI CFI RMSEA Indirect
effect
Lower
bound
CI
Upper
bound
CI
116
1 .048 .945 .899 .921 .085 .467 -.041 .121
2 .037 .958 .914 .944 .057 .085 -.090 -.002
3 .091 .952 .897 .916 .092 .006 -.111 -.025
4 .049 .952 .897 .917 .092 .020 -.092 -.016
5 .042 .962 .935 .957 .064 .360 -.032 .008
6 .035 .966 .936 .968 .045 .053 -.059 .149
7 .074 .971 .941 .960 .065 .005 -.075 -.016
8 .036 .975 .946 .966 .060 .017 -.045 -.006
9 .094 .889 .842 .875 .084 .263 -1.496 .235
10 .087 .880 .821 .853 .091 .204 -2.653 .211
Note:
Indirect effect or mediation is significant at the 0.05 level*; 0.01 level ** and 0.001***
Model 11 is combined model with both predictors (individualism and collectivism) along with all the mediators run
together
Ind = Individualism; col = collectivism
Overall the results suggest that in general the single mediator models confirms a better
mediating effect on the relationship between the cultural orientation and group identity as when
the mediators are combined the model fit indices reduced. In general, it can be seen that for both
the cultural orientation, i.e., for both collectivists and individualists the number of days and time
members invests has a full mediating effect on group identification. This further confirms that
increase in frequency of Facebook use increased the group identification not the type of activities
members indulged in.
117
H2: Group identification would be stronger among members from collectivist cultures than
members from individualist cultures.
An independent sample t-test for group identification confirmed that Indonesian participants
identified with their group members more than the UK participants and the difference was
significant (t = -4.08, df = 430, p < 0.05, two tailed and had a weak effect size was (d = 0.25) which
supports H2.
To summarize, the results confirmed that the higher the consistency of being available on SNS,
like Facebook, the higher members would be able to identify with others. The activities on
Facebook did not have any effect on group identification. Furthermore, members from collectivist
countries identified with their group members more than the members from individualist countries.
5.3.21.2. Online Self-Enhancement
This model looked at the impact of cultural orientation (individualism/ collectivism) on online
self enhancement on Facebook. Self-enhancement was measured using 7 collectivist traits and 7
individualist traits. Therefore, it was decided that self enhancement be measured on the cultural
traits separately. Hence at first the single mediator models were run in AMOS followed by the
combined mediator models. The aim was to measure how participants with particular cultural
orientation (collectivism/ individualism) scored on the cultural orientation traits, i.e., how much
they self-enhanced on their particular cultural traits. The results will be discussed to evaluate the
hypothesis:
H3: Forms of online self-enhancement will vary across cultures. Individualists should rate
themselves higher on individualist attributes compared to others online whereas collectivists
should rate themselves higher collectivist attributes compared to others.
An independent sample t-test for self-enhancement confirmed that participants in the UK scored
themselves higher on collectivist traits than the participants in Indonesia (UK: M = .93, SD = .702;
Indonesia: M = .40, SD = .574), t = .873, df = 430, p < 0.05, two tailed and had a large effect size
(d = .83). This was certainly not as expected as members were anticipated to score themselves
higher on their cultural specific traits thus it failed to support H3. The results also confirmed that
118
participants in the UK scored themselves higher on individualist traits than participants in
Indonesia (UK: M = .31, SD = .780, Indonesia: M = .17, SD = .493), t = 2.251, d = 430, p < 0.05,
two tailed and had a weak effect size (d = .21). This was as per what was expected as members
from individualist cultures (UK) would be expected to score higher on individualist traits than
members from collectivist cultures which supports H3. However as individualist members (UK)
scored higher on collectivist traits than the members from collectivist country (Indonesia), overall
it seems to partially support H3 as the effect size was larger for this difference in scores than when
members scored themselves higher on cultural specific traits.
H4: Facebook activities and Facebook frequency will mediate the relationship between
cultural orientation of participants and their online self-enhancement such that increase in
Facebook activities and FB frequency will increase self-enhancement in both cultures.
Figure 5.6 illustrates the single mediator models where scores on collectivism was used as the
IV and self-enhancement was measured on collectivist traits as the DV with the single mediators
each time. The results confirm that both active (b = .000, p < 0.05, .085, .307) and passive (b =
.003, p < 0.05, .065, .303) type of Facebook activities significantly mediated the relationship
between cultural orientation (collectivism traits) and self-enhancement such that increase in
collectivism decreased Facebook activities (both active and passive). However, increase in
Facebook activities decreased online self-enhancement failing to support H4. This was also the
case when individualism was used as the IV, both active (b = .000, p < 0.05, .045, .178) and passive
(b = .003, p < 0.05, .065, .303) Facebook activities significantly mediated the relationship between
individualism and self-enhancement (see Figure 5.8). The results confirm that increase in
individualism decreases Facebook activities (both active and passive) and increase in Facebook
activities (both active and passive) decreases self-enhancement which fails to support H4.
The other mediators (FB Days: number of days of Facebook use and FB HR: number of hours
invested on Facebook) failed to show any significant effect on self-enhancement. This confirms
that when cultural orientation increased it resulted in decrease in use of SNS, like Facebook. This
could be attributed to social restrictions, accessibility and availability of SNS or might also be
attributed to cultural norms. On the contrary when members used SNS, like Facebook it failed to
help them self-enhance which was not what was expected and thus failed to support H4. In order
119
to prevent complexity only significant mediating effect has been highlighted. There also existed a
fairly expectable model fit indices for all the models had fairly acceptable model fit indices (see
Table 5.11 below) confirming that the results were representative of the population in both
countries.
Figure 5.6.
Models 1 to 4 Illustrates Standardized Path Coefficients of the Mediating Effect on Online Self-
Enhancement using Collectivism as the predictor variable on Collectivist Attributes
Note: FB = Facebook; FB Active = active interaction on Facebook; FB Passive = passive interaction on Facebook; FB Days = average number of days’
respondents had been on Facebook in the last 14 days and FB HR = average number of time (in minutes) respondents had been on Facebook in
the last 14 days
120
Figure 5.7.
Models 5 to 8 Illustrates Standardized Path Coefficients of the Mediating Effect on Online Self-
Enhancement using Collectivism as the predictor variable on Individualist Attributes
Note: FB = Facebook; FB Active = active interaction on Facebook; FB Passive = passive interaction on Facebook; FB Days = average number of days’
respondents had been on Facebook in the last 14 days and FB HR = average number of time (in minutes) respondents had been on Facebook in
the last 14 days
121
Figure 5.8.
Models 9 to 12 Illustrates Standardized Path Coefficients of the Mediating Effect on Online Self-
Enhancement using Individualism as the predictor variable on Collectivist Attributes
Note: FB = Facebook; FB Active = active interaction on Facebook; FB Passive = passive interaction on Facebook; FB Days = average number of days’
respondents had been on Facebook in the last 14 days and FB HR = average number of time (in minutes) respondents had been on Facebook in
the last 14 days
122
Figure 5.9.
Models 13 to 16 Illustrates Standardized Path Coefficients of the Mediating Effect on Online
Self-Enhancement using Individualism as the predictor variable on Individualist Attributes
Note: FB = Facebook; FB Active = active interaction on Facebook; FB Passive = passive interaction on Facebook; FB Days = average number of
days’ respondents had been on Facebook in the last 14 days and FB HR = average number of time (in minutes) respondents had been on Facebook
in the last 14 days
Figure 5.10 (see Appendix 8) illustrates the combined models where all the four mediators (FB
active, FB Passive, FB Days and FB HR) were run together in the same model using cultural
orientation (collectivism/ individualism) as the IV and self enhancement on collectivist attributes
as DV. The aim was to understand whether cultural orientation had an impact on self-enhancement
on collectivist attributes mediated by the four mediators. While the combined model failed to
123
show any significant mediating effect on the relationship between collectivism cultural orientation
and self enhancement (b = .097, p > 0.05, .034, 11.779) (see Appendix 8, Model 17), individualism
cultural orientation on the other hand showed significant overall mediating effect (b = .001, p <
0.05, 3.531, 48.288) (see Appendix 8, Model 18). The model confirms that increase in
individualism decreases Facebook activities (Active: b = -.17, p < 0.05; Passive: b = -5.22, p <
0.05, FB Days: b = -27.40, p < 0.05 and FB HR: b = -13.59, p < 0.05). The path from FB Days
and FB HR are significant and further confirms that increase in the number of days one uses
Facebook (FB Days) decreased self enhancement (b = -.11, p < 0.05) and increase in the time
invested on Facebook use (FB HR) also significantly decreases self-enhancement (b = -.20, p <
0.05) which partially supports H13. The path from individualism to self-enhancement was also
significant suggesting that there is also a direct effect of cultural orientation on self enhancement
(b = -.17.50, p < 0.05). However, as the direct path from individualism to self enhancement is
negative, it suggests that increase in individualism (cultural orientation) has a negative effect on
self-enhancement, such that increase in cultural individualism will decrease self-enhancement on
collectivist traits online and was supported by acceptable model fit indices (see Table 5.11). This
result could be suggesting that there exists a correlation between the four mediators used as the
combined model showed a significant mediation effect on self-enhancement for individualism
cultural orientation. Furthermore, when the mediators (FB Days; number of days used Facebook;
FB HR: number of hours invested on Facebook) are analysed separately it failed to have any
significant mediating effect on both cultural orientations, which further supports the analysis that
if self-enhancement, needs to understood in-depth, all the four mediators need to analysed together.
Figure 5.11 (see Appendix 9) illustrates the combined models where all the four mediators (FB
active, FB Passive, FB Days and FB HR) has been run together in the same model using cultural
orientation (collectivism/ individualism) as the IV and self enhancement on individualist attributes
as DV. The aim was to understand whether cultural orientation had an impact on self-enhancement
on individualist attributes mediated by the four mediators. The results confirmed that Model 19 (b
= .404, p > 0.05, -6.439, 2.334) and Model 20 (b = .305, p > 0.05, -3.228, 26.331) which used
individualist attributes for self-enhancement as DV both failed to show any significant mediating
effect on self-enhancement hence has not been discussed further.
124
H5: There will be a positive relationship between number of Facebook friends’ participants
have in both cultures and online self enhancement such that increase in number of social
contacts on Facebook will increase feelings of self-enhancement for participants in both UK
and Indonesia.
First, tests for differences between countries were carried out. Independent samples t-tests for
perceived online self-enhancement and the total number of friends, members had on Facebook
showed significant differences for both variables: Indonesian participants had a higher number of
friends on Facebook (M = 47.45, SD = 72.42) than participants in UK (M = 25.83, SD = 24.17, t
= -4.08, df = 424, p < 0.05, d = .40). Indonesian participants scored lower level of self-
enhancement (M = .23, SD = .57) than the UK participants (M = .63, SD = .91, t = 8.04, df = 424,
p < 0.05, d = .53).
In order to take account of the country differences, separate correlations between number of friends
and self-enhancement were calculated for both samples. Again, this failed to support H5 because
correlations were nil for both Indonesians (r = .00) and UK participants (r = .00).
Table 5.11.
Illustrates the Model Fit Indices, the Mediation Effect (Indirect Effect) & the Confidence
Intervals for all the models for Online Self-Enhancement
Model RMR GFI NFI CFI RMSEA Indirect
effect
Lower
bound
CI
Upper
bound
CI
1 .052 .962 .907 .938 .063 .000 .085 .307
2 .053 .964 .897 .943 .050 .003 .065 .303
3 .118 .964 .893 .974 .079 .059 .012 .236
125
4 .104 .954 .800 .831 .093 .252 -.030 .223
5 .065 .970 .924 .957 .053 .648 -.082 .037
6 .061 .972 .924 .970 .037 .821 -.068 .094
7 .150 .969 .880 .014 .070 .238 -.159 .023
8 .063 .969 .881 .914 .070 .250 -.168 .031
9 .048 .978 .949 .981 .035 .000 .045 .178
10 .053 .964 .897 .943 .050 .003 .065 .303
11 .118 .981 .930 .966 .044 .067 .005 .133
12 1.00 .971 .888 .921 .069 .139 -.003 .098
13 .071 .958 .911 .939 .066 .743 -.032 .045
14 .067 .956 .894 .932 .059 .292 -.015 .080
15 .146 .952 .846 .871 .096 .931 -.061 .046
16 .070 .957 .860 .887 .088 .728 -.046 .025
17 .111 .894 .825 .862 .081 .097 .034 11.779
18 .111 .882 .797 .833 .090 .001 3.531 48.288
19 .115 .902 .839 .878 .077 .404 -6.439 2.334
20 .111 .874 .791 .825 .094 .305 -3.228 26.331
Note:
Indirect effect or mediation is significant at the 0.05 level*; 0.01 level ** and 0.001***
Models 1 to Models 16 are the single mediator models (see Appendix 14, 15, 16 and 17)
Model 17, Model 18, Model 19 and Model 20 are combined mediator models with both predictors (individualism and
collectivism) along with all the mediators run together
Ind = Individualism; col = collectivism
126
Summary
Overall the results suggest that in general the single mediator models confirmed a better
mediating effect on the relationship between the cultural orientation and self-enhancement as when
the mediators were combined the model fit indices reduced. In general, it can be seen that for both
the cultural orientation, i.e., for members from both cultural orientation activities on Facebook
members indulge in has a partial negative mediating effect on self enhancement, such that increase
in the active activities and increase in passive activities on Facebook decreased self-enhancement.
However, when the mediators were combined in a single model increase in cultural orientation
decreased Facebook activities (active, passive) and number of days and time participants invested
on Facebook. Furthermore, increase in individualism cultural orientation decreased self-
enhancement on collectivist attributes further confirming that members self-enhance on cultural
specific attributes. The results also confirmed that members from both countries scored themselves
higher on collectivist attributes when compared with their social contacts on Facebook. While this
is something that was expected for the members from collectivist countries but not for members
from individualist country as it was anticipated that they would score higher on individualist
attributes. This further explains how the level of anonymity on SNS shifts focus of members from
personal self to social self.
5.3.21.3. Online Perceived Social support
This model looked at the impact of cultural orientation (individualism/ collectivism) on how
much members perceived the level of social support they received on SNS, like Facebook. While
Models 1,2,3,4 (Figure 5.12) and Model 5,6,7,8 Figure (5.13) are the single mediator models and
Figure 5.14 (see Appendix 10) combined models where all the four mediators are analysed together
for each cultural orientation. Through these models, the below hypotheses are being evaluated:
H6: Facebook activities (time and frequency) would mediate the relationship between
culture and social support, such that the increase in user behaviour (Facebook activities and
Facebook frequency) would increase perceived social support for its members in both
cultures.
127
Figure 5.12. Demonstrates that increase in collectivism reduces Facebook activities (FB Active:
beta = -.63, FB Passive: beta = -.53, FB Days: beta = -4.14 and FB HR: beta = -2.24). However,
increase in Facebook activities was seen to increase perceived social support members received
online for collectivism (FB Active: beta = .34, FB Passive: beta = .58, FB Days: beta = .07 and FB
HR: beta = .11).
Figure 5.12.
Models 1 to 4 Illustrates Standardized Path Coefficients of the Mediating Effect on Perceived
Online Social Support using Collectivism as the predictor variable
128
Note:
FB = Facebook; FB Active = active interaction on Facebook; FB Passive = passive interaction on Facebook; FB Days = average number of days’
respondents had been on Facebook in the last 14 days and FB HR = average number of time (in minutes) respondents had been on Facebook in
the last 14 days
Figure 5.13.
Models 5 to 8 Illustrates Standardized Path Coefficients of the Mediating Effect on Perceived
Online Social Support using Individualism as the predictor variable
Note:
FB = Facebook; FB Active = active interaction on Facebook; FB Passive = passive interaction on Facebook; FB Days = average number of days’
respondents had been on Facebook in the last 14 days and FB HR = average number of time (in minutes) respondents had been on Facebook in
the last 14 days
129
Figure 5.13 also demonstrates that increase in levels of individualism decreased Facebook
activities (FB Active: beta = -.47, FB Passive: beta = -.34, FB Days: beta = -2.76 and FB HR: beta
= -.90) suggesting that increase in the levels of both cultural orientations reduce the use of SNS
like Facebook. While Models 1234, see Figure 5.12 and Models 5,6 and 7 (see Figure 5.13)
illustrates partial mediating effect of Facebook activities on perceived social support, Model 8 (see
Figure 5.13) illustrates a full mediating effect of the number of hours’ members use SNS (FB HR)
on perceived social support. It is however seen that the decrease in Facebook activities is higher
in collectivism than in individualism. The Model fit for all the models suggest that the models
could be accepted as it suggest that the reproduced models matches with the original sample.
These results support H6 as it is seen that the use of SNS, like Facebook helps members perceive
that they receive social support online.
Figure 5.14. (see Appendix 10) demonstrates a combined model with all the mediators in study
(FB Active, FB Passive, FB Days and FBHR). The figure demonstrates that when all the mediators
are combined in the same model the Model fit seems to decrease. Model 9 demonstrates that the
direct path FB active to support (beta = -.09, p>0.05) and FB Passive to support (beta = 1.11, p >
0.05) has now become non-significant. The results in Model 9 also shows that the relationship
between one’s cultural orientation and perceived social support is partially mediated by the number
of days (FB Days) one uses FB and the number of hours’ participants spent on FB.
Model 10 also predicted a significant mediation effect with all the mediators combined.
However, the path from FB Active to support (beta = -.07, p > 0.05), FB Passive to support (beta
= 1.34, p > 0.05) and FBHR to support (beta = .30, p > 0.05) were now non-significant.
Interestingly the direct path from Individualism to Support is also non-significant (beta = 1.52, p
> 0.05) confirming full mediation. Unfortunately, all the Model fit indices confirm poor model fit
as all the fit indices were above the acceptable range (see Table 5.12).
130
Table 5.12.
Illustrates the Model Fit Indices, the Mediation Effect (Indirect Effect) & the Confidence
Intervals for all the models for Perceived Online Social Support
Model RMR GFI NFI CFI RMSEA Indirect
effect
Lower
bound
CI
Upper
bound
CI
1 .066 .947 .943 .961 .069 .000 -.185 -.066
2 .057 .950 .938 .962 .058 .000 -.481 -.184
3 .189 .945 .940 .953 .087 .000 -.227 -.117
4 .092 .951 .945 .958 .081 .000 -.191 -.090
5 .058 .959 .957 .974 .057 .001 -.132 -.044
6 .052 .956 .949 .972 .050 .003 -.167 -.043
7 .180 .953 .951 .964 .077 .000 -.149 -.062
8 .082 .962 .960 .974 .065 .003 -.091 -.019
9 .139 .887 .876 .905 .079 .001 -2.913 -.896
10 .133 .875 .858 .887 .087 .002 -3.449 -.818
Note:
Indirect effect or mediation is significant at the 0.05 level*; 0.01 level ** and 0.001***
Model 11 is combined model with both predictors (individualism and collectivism) along with all the mediators run
together
Ind = Individualism; col = collectivism
131
H7: Increase in the number of social contacts on Facebook would have a positive impact on
perceived social support that members receive online.
First, tests for differences between countries were carried out. Independent samples t-tests for
perceived online social support and the total number of friends’ members had on Facebook showed
significant differences for both variables: Indonesian participants had a higher number of friends
on Facebook (M = 47.45, SD = 72.42) than participants in UK (M = 25.83, SD = 24.17, t = -.078,
df = 424, p < 0.05, d = .40). However, Indonesian participants scored lower on perceived social
support (M = 2.72, SD = .83) than the UK participants (M = 3.42, SD = .98, t = 8.04, df = 430, p
< 0.05, d = 1.27). Social support scores also differed in Study 1 as UK participants scored higher
on perceived social support than participants in Indonesia (see Study 1, sub-section 4.33). For the
overall sample, these findings would support the idea that, contrary to H7, a higher number of
friends indicates lower levels of perceived social support. In order to take account of the country
differences, separate correlations between number of friends and social support were calculated
for both samples. Again, this failed to support H7 because correlations were close to zero and non-
significant for Indonesian (r = .04) and UK (r = .06) participants.
Overall it seems that the activities on Facebook (both type of activities and time investment)
helps members to get social support online. While it is important to be present on SNS to develop
the relationship with other members on the social network thereby helping to increase the size of
social network one has, it is evident that there exists cultural influence on such differences in the
network size. Maintaining an acceptable and genuine network size has also been highlighted.
While the single mediation models illustrated acceptable model fit indices, it reduced when the
mediators were combined. The combined model where collectivism was used as a predictor
variable supported the indirect effect of Facebook activities, which highlights that there might exist
a close relationship between the mediators due to which the effect changes when analysed
separately and when combined.
5.3.21.4. Online Trust
This model looked at the impact of cultural orientation (individualism/ collectivism) on trust
members had for their social contacts on Facebook. All the model fit indexes were within the
132
acceptable range suggesting that the models are representative of the original sample (see Table
5.13). Figure 5.15 demonstrates all the single mediator models (1,2,3,4) where collectivism was
used as the predictor variable and Figure 5.16 demonstrates all the single mediator models (5,6,7,8)
where the models were each time to gauge an understanding of the effect of mediators on the on
online trust. The results below will be analysed under each hypothesis:
H8. Activities and the frequency of Facebook use will mediate the relationship between
culture and online trust, such that higher use of Facebook (activities and frequency) will
increase online trust in both cultures.
Results for the single mediator models (Figure 5.15 and 5.16) suggests that increase in use of
Facebook had a positive impact on online trust scores. However not all the models had a significant
mediation effect on online trust. Type of activities and time invested on Facebook had an indirect
effect on online trust score, Facebook Active (beta = .08, p < 0.05), Facebook Passive (beta = .12,
p < 0.05) and hours of Facebook (FB HR: beta =.02, p <0.05, p = < 0.05), however the path from
Facebook HR (FB Days: beta = .02, p > .05, CI: =.085, -.006) further confirming that the mediating
effect was not strong which was supported by the fit indices as while RMR, GFI, NFI and CFI
were fairly within acceptable range, RMSEA was high.
133
Figure 5.15.
Models 1 to 4 Illustrates Standardized Path Coefficients of the Mediating Effect on Online Trust
with Collectivism as the predictor variable
Note: FB = Facebook; FB Active = active interaction on Facebook; FB Passive = passive interaction on Facebook; FB Days = average number of days’
respondents had been on Facebook in the last 14 days and FB HR = average number of time (in minutes) respondents had been on Facebook in
the last 14 days
When individualism was used as a predictor variable Facebook Active (beta = .07, p < 0.05 and
Facebook Passive (beta = .09, p < 0.05) further confirmed that the type of activities members
indulged in had a significant mediating effect on trust scores. This confirms that the results partially
134
support H8, as while type of activities had an indirect effect on the level of trust members had for
their members on SNS, like Facebook, however the frequency of Facebook use, i.e., the number
of hours and number of days invested on SNS like Facebook does not have an impact on managing
the level of trust among members online.
Figure 5.16.
Models 5 to 8 Illustrates Standardized Path Coefficients of the Mediating Effect on Online Trust
with Individualism as the predictor variable
Note: FB = Facebook; FB Active = active interaction on Facebook; FB Passive = passive interaction on Facebook; FB Days = average number of
days’ respondents had been on Facebook in the last 14 days and FB HR = average number of time (in minutes) respondents had been on Facebook
in the last 14 days
135
Model 9 (see Appendix 11) demonstrates that when all the four mediators were combined in
the same model the direct effect from collectivism to trust which was significant in all the single
models (Model 1, 2, 3 and 4; see Figure 5.15) now becomes non-significant (beta = .91, p > 0.05).
The direct effect from FB Active to Trust (beta = .03, p > 0.05) and FB Passive to Trust (beta =
.57, p > 0.05) which was significant before also became non-significant. However, the direct effect
from collectivism to all the mediators was still statistically significant and had higher negative
effect suggesting that when all the mediators are combined increase in collectivism reduces FB
activities (FB Active and FB Passive) and also reduces the time members invest on Facebook (FB
Days and FBHR) more than when the mediators are analysed separately. Similar effect was also
seen in Model 10 where individualism was used as a predictor variable. An indirect effect was
tested using a bootstrap estimation with 2000 samples (Shrout & Bolger, 2002) which indicated
that the indirect coefficient was significant, beta = .013, SE 90 % confidence interval = -3.674, -
.471. However, when collectivism was used as a predictor using the same method, there was no
statistically significant mediation effect found, beta = .058, p > 0.05.
The results further confirm as unlike the single models where FB activities seems to have
significant mediating effect (see Table 5.13) for both collectivism and individualism cultural
orientation, however when the mediators were combined a significant mediating effect was only
seen for individualism cultural orientation. Model 10 confirms that although the direct effect from
Facebook activities becomes non-significant, type of activities and time members spend on
Facebook does seem to mediate the relationship between individualism and online trust. Although
increase in individualism in members seem to decrease online activities and time spent on
Facebook. Therefore, confirming full mediation caused by the presence of the four mediators (FB
Active, FB Passive, FB Days and FBHR) on the relationship between individualism and online
trust for its members. Contrary to the single mediation model, Facebook activities reduced level
of trust for members online, however the path was quite weak (beta = -.01, p > 0.05). The above
results are also supported by acceptable model fit indexes (see Table 11). It is however seen that
as more meditators were added in the model which increased the number of parameters, it reduced
the model fit indexes (see Table 5.13).
136
Table 5.13.
Illustrates the Model Fit Indices, the Mediation Effect (Indirect Effect) & the Confidence
Intervals for all the models for Online Trust
Model RMR GFI NFI CFI RMSEA Indirect
effect
Lower
bound
CI
Upper
bound
CI
1 .047 .959 .938 .961 .060 .005 -.079 -.014
2 .038 .966 .940 .972 .042 .009 -.108 -.019
3 .120 .957 .926 .945 .079 .187 -.095 .008
4 .060 .957 .927 .946 .078 .048 -.085 -.006
5 .036 .977 .966 .989 .033 .011 .011 -.059
6 .032 .975 .957 .989 .027 .038 .038 -.064
7 .102 .977 .961 .979 .049 .296 .296 -.042
8 .042 .981 .968 .988 .037 .203 .203 -.034
9 .101 .897 .863 .897 .076 .058 -1.961 -.098
10 .093 .889 .843 .877 .083 .013 -3.674 -.471
Note:
Indirect effect or mediation is significant at the 0.05 level*; 0.01 level ** and 0.001***
Model 11 is combined model with both predictors (individualism and collectivism) along with all the mediators run
together
Ind = Individualism; col = collectivism
137
H9: Levels of trust would differ due to cultural differences such that collectivists would score
higher on the level of trust for their members than individualists.
An independent sample t-test was run on the trust scores between UK and Indonesia which
confirmed that the UK participants scored higher on the trust scale (M = 3.14, SD = .69) than the
Indonesian sample (M = 3.09, SD = .69) as predicted however the difference in scores was not
statistically significant (t = .694, df = 430, p > 0.05, two tailed) thereby failed to support H9. On
the contrary Study 1 (see sub section 4.33) confirmed that there was a statistically significant
difference in level of trust members had for their social contacts online as anticipated Indonesian
participants scored higher on their level of trust online for their social contacts than UK
participants.
To summarize, the above results demonstrates that while the single mediator models illustrates
that the level of trust increased when mediated by the type of activities members indulged in, i.e.,
the activities on Facebook seem to help increase the level of trust for its members on Facebook.
While the combined mediator model showed that there was a mediating effect on the level of trust
when all mediators were combined, the direct path were non-significant. It could be that as
highlighted that the mediators might be closely co-related due to which when combined, the effect
changed. While Study 1 confirmed that members from collectivist countries did trust their social
contacts online more than members from individualist countries, this difference could not be
established.
5.3.21.5. Online Privacy Concern
This model looked at the impact of cultural orientation (individualism/ collectivism) on how
much members were concerned about their privacy when they communicate online. Figure 5.17.
demonstrates all the models (1, 2, 3, 4) which consists of collectivism as the predictor variable and
privacy as the outcome variable and Figure 5.18. demonstrates all the models (5,6,7,8) which
consists of individualism as the predictor variable and privacy as the outcome variable. Their
relationship was mediated by the presence of a third variable which are the mediators (FB Active,
FB Passive, FB Days and FB Time). The results were analysed under each hypothesis.
138
H10: There would be a negative relationship between Facebook activities (active & passive),
Facebook frequency (days and hours spent) and online privacy concern, such that increase
in Facebook activities and Facebook frequency would decrease privacy concern among
online users.
When collectivism was used as a predictor variable (see Figure 5.17) in the single mediator
models, only passive activities (FB Passive: beta = .22, p < 0.05) on Facebook mediated the
relationship between cultural orientation and privacy concern and a full mediation effect was
achieved (b=0.18, p < 0.05) which failed to support H10.
Figure 5.17.
Model 1 to 4 Illustrates Standardized Path Coefficients of the Mediating Effect on Online
Privacy Concern using Collectivism as the predictor variable
139
Note: FB = Facebook; FB Active = active interaction on Facebook; FB Passive = passive interaction on Facebook; FB Days = average number of days’
respondents had been on Facebook in the last 14 days and FB HR = average number of time (in minutes) respondents had been on Facebook in
the last 14 days
Whereas, individualism cultural traits seemed to demonstrate that Facebook activities (Active:
b = .11, p < 0.05; Passive: b = 0.02, p < 0.05) both has a significant mediating effect on privacy
concern on Facebook. Additionally, number of days’ members spent on FB (FB days: b = .41, p <
0.05) also had a significant mediating effect on privacy concern and for all the models a partial
mediation effect was achieved. Overall it was seen that increase in Facebook activities and number
of days’ members went on Facebook increased privacy concern among its users and also failed to
support H10.
140
Figure 5.18.
Model 5 to 8 Illustrates Standardized Path Coefficients of the Mediating Effect on Online
Privacy Concern using Individualism as the predictor variable
Note: FB = Facebook; FB Active = active interaction on Facebook; FB Passive = passive interaction on Facebook; FB Days = average number of days’
respondents had been on Facebook in the last 14 days and FB HR = average number of time (in minutes) respondents had been on Facebook in
the last 14 days
The combined model (see Figure 5.19, Appendix 12) demonstrates in both the model which
tests both cultural orientation, suggests that increase in collectivism or individualism decreases
Facebook activities (Active: b = -4.30, p < 0.001; Passive: b = -3.11, p < 0.001) and Facebook
frequency (FB Days: -17.19; p < 0.001; FBHR: b = -8.80, p < 0.001) such that the more collectivist
or individualist one is the lesser they would be using Facebook. When collectivism was used as
the predictor variable, the indirect effect non-significant, b = .369, p > 0.05 but the indirect effect
141
was statistically significant b = .002, p < 0.01, when individualism was used as the predictor
variable. This further suggests that individualism was associated with approximately .002 increase
in privacy concern as mediated by FB activities and FB frequency. Although the path from
mediators to the outcome variable (DV) was not significant in the combined models (Model 9 &
Model 10) which tests all the four mediators at the same time are not significant, the single
mediator models which tests each mediator one by one (Models 2, 5, 6,7, see Figure 5.17 and
Figure 5.18) confirms that FB activities (passive) where members are more inclined in passively
looking through their Facebook contacts for collectivist orientation (b = .018, p < 0.05) & for
individualist orientation both (active: b = .011, p < 0.05 and passive; b = .002, p < 0.05) activities,
where members indulge in not just spying around the activities of their social contacts but also
interacting with their social contacts. Additionally, increase in the number of days’ members use
Facebook (FB Days) increases concern for online privacy on Facebook (b = .042; p < 0.05), which
again failed to support H10. The model fit indexes however is reduced with the inclusion of all the
mediators in one model (see Table 5.14). All the model fit indexes have gone above the acceptable
fit range and this fit range is further compromised when both the cultural constructs (individualism
and collectivism) are entered together in one model. The single mediator models however all had
acceptable fit ranges. Model 11 which combined all both the cultural orientation (individualism
and collectivism) in one model along with the four mediators has not been discussed further as it
failed to meet the model fit indices (see Table 5.14).
Table 5.14.
Illustrates the Model Fit Indices, the Mediation Effect (Indirect Effect) & the Confidence
Intervals for all the models for Online Privacy Concern
Model RMR GFI NFI CFI RMSEA Indirect
effect
Lower
bound
CI
Upper
bound
CI
142
1 .060 .951 .905 .925 .087 .143 .003 -.067
2 .050 .959 .911 .941 .063 .018 -.128 -.020
3 .134 .952 .875 .891 .110 .533 -.064 .027
4 .067 .951 .877 .894 .108 .907 -.054 .038
5 .048 .978 .960 .981 .044 .011 -.061 -.010
6 .044 .973 .944 .975 .042 .002 -.109 -.018
7 .112 .984 .962 .980 .048 .041 -.069 -.006
8 .044 .990 .974 .994 .026 .256 -.038 .004
9 .108 .886 .835 .865 .090 .369 -1.757 .329
10 1.00 .878 .815 .844 .097 .002 -4.562 -.811
Note:
Indirect effect or mediation is significant at the 0.05 level*; 0.01 level ** and 0.001***
Model 11 is combined model with both predictors (individualism and collectivism) along with all the mediators run
together
Ind = Individualism; col = collectivism
H11: Privacy concern will be more explicit among members from collectivist cultures than
members from individualist cultures, such that collectivists would score higher in their
privacy concern than individualists.
An independent sample t-test was run on participants scores on online privacy concern between
countries and the results confirmed that Indonesian participants scored higher on online privacy
score (M = 3.11, SD = .55) than participants in UK (M = 2.97, SD = .80), t = -2.17, df = 430, p <
0.05, two tailed however it had a weak effect size, (d = .20) suggesting that the members from
collectivist countries ae more concerned than the members from individualist countries which
143
supports H11. However, as the difference in mean is quite small (Mean difference = .14), which
probably effects the effect size.
To summarize, the results confirms that increase in cultural orientation whether individualism
or collectivism decreases use of Facebook both in terms of activities and frequency of use which
meet. As members from collectivist countries prefer maintaining their group identity, it reflects on
their ratings on privacy concern as they were slightly more concerned about their online privacy
than the members from individualist countries. However, the results also confirmed that the
members from individualist countries also showed concern about their privacy when they used
SNS. Facebook activities and the number of days’ members spent on SNS, like Facebook increase
the privacy concern for its members as when members invest their time to interact through the
various activities they indulge in with their social network on SNS, they get personal information
from other members which helps to predict their behaviour in advance. As members from
collectivist countries are more cautious they become more concerned about maintaining their
privacy online.
5.9. Summary of Results
a) Online Group Identity
The results suggest that irrespective of whether members were collectivist or individualist, i.e.,
whether they scored higher on collectivist attributes or individualist attributes, the time spent
members invested on Facebook helped members in group identification not the type of activities
members are indulged in. This could be explained by the fact that when members spend time
interacting with their social group, it would help them to identify themselves with other members.
It could also be argued that when members become part of a group, their availability on SNS helps
others in the group to evaluate their loyalty and commitment towards the group. The results also
confirmed that Indonesian participants identified with their group members more than the UK
participants, which was expected as their social network on SNS would generally consist of a
higher number of social contacts with whom they might have established offline relationships.
144
b) Online Self-Enhancement
The results confirm that activities on SNS, like Facebook reduced feeling of self-enhancement
for members from both cultural orientations. This could further be explained when members invest
more time on Facebook, they end up consuming more information about others on Facebook and
this might have a negative impact on their self-perception as members might start comparing their
personal achievements in comparison to others. The results further confirm that members on
Facebook self-enhance on cultural specific traits as it was seen that increase in individualism
decreased self-enhancement on collectivist attributes. However, results also confirmed that
participants in UK scored themselves higher on individualist traits than the Indonesian participants
which was expected, Interestingly, UK participants also scored themselves higher on collectivist
attributes when compared with others. This could be due to the cultural specific behaviour as
members from collectivist country (Indonesia) believe in being humble whereas members from
individualist (countries) believe in showcasing their positive attributes more openly. Additionally,
it could also be due to the fact that as SNS, involves more collective behaviour, there seems to be
a shift from personal shelf to the collective self as has been supported by SIDE.
c) Perceived Online Social Support
The results confirmed that use of SNS, like Facebook and investing time to interact with others
not just opened the potential of increasing the network size but it also increased the potential of
being able to access support in an online environment from a wider population which was not
possible before. This was also supported by significant mediation effect which confirmed that
Facebook activities (active and passive) and Facebook frequency (number of days and the number
of time one invests) on Facebook has a positive mediating effect on relationship between culture
and perceived social support online. This highlights how important SNS has become and thus
could also be utilized as a platform where members receive social support from their social group.
d) Online Trust
The single models confirmed that for both cultural orientation increase in FB activities
increased trust for their members. While the combined mediator model showed a significant
mediation effect, activities on Facebook did not have any direct effect. This could be due to the
existence of correlation with the other mediators which resulted in a significant mediation effect.
145
Although UK participants scored higher than the Indonesian participants on the trust scale as
predicted, the difference in scores was not statistically significant.
e) Online privacy concern
The results suggest that increase in Facebook activities (active and passive) and the number of
days’ members used Facebook increased privacy concern on Facebook. While collectivism
cultural orientation seemed to only have an effect on online privacy concern mediated by passive
activities on Facebook, i.e., by passively using Facebook, members with collectivist cultural
orientation would become more concerned about their privacy online. Additionally, members from
individualist cultural orientation would become concerned about their privacy online by interacting
actively as well as passively on Facebook. This can be explained as members on Facebook increase
their interaction on Facebook, they get to know their social contacts more, which makes them more
concerned about their privacy online. As SNS is a platform where members can choose their social
contacts and form their social groups, members might not always be comfortable in sharing
information comfortably to all group members. As members from collectivist countries prefer
maintaining their group identity, it reflects on their ratings on privacy concern as they were slightly
more concerned about their online privacy than the members from individualist countries.
However, the results also confirmed that the members from individualist countries did show
concern about their privacy when they used SNS
It is important to point out that while attempt was made to use the combined models in most
cases, however due to the poor model fit indices, it was decided that results from the single
mediator models be included in the analysis. For all the constructs, the single models had a better
model fit than the combined models. The next chapter will now discuss the “Study 3” which was
the experimental study (scale development) that helped to further refine “Study 4” which was the
experimental study.
146
Chapter 6
Study 3: Experiment (Operationalization)
6.1. Rationale for this study
The role of culture is important in the development of one’s self-construal and how members
perceive the environment around them, which also has an impact on their decisions and
behaviours. While members from collectivist countries will have a self-construal with collectivist
attributes that promotes interdependence, members from individualist countries will develop a
self-construal with individualist attributes that promote independent attributes (Hofstede, 1980;
Markus & Kitayama, 1991; Oyserman & Lee, 2008). Therefore, being able to understand the
change in cultural behaviour could be achieved by manipulating the cultural selves which could
be achieved by priming (Gardner et al., 1999). While the previous two chapters were based on
responses from online surveys, this chapter and the next will present the results and analysis of the
priming experiments that were conducted across the UK, India and Indonesia. In order to better
assess the causal impact of cultural orientation it was important to pursue experiment-based
evidence that would help to justify the causal role assigned to cultural self-construal in the analyses
of the survey data. Furthermore, while the online survey results are determined by the variability
in responses that participants brought with them. the experiments further helped to measure the
causal strength of cultural factors by directly influencing their variability (see Figure 2.2 Visual
Illustration for the Research Overview) which illustrates the research framework for this thesis. A
paper-and-pencil method of experimentation was used as this helped to gain access to the different
populations and was simple to administer. This chapter will discuss the ‘operationalization stage’
during which the focus was on the development of the priming manipulation and also collect
participant feedback on the overall participant experience. As such, the present chapters can
present preliminary findings only. The next chapter (Chapter 7) will then discuss ‘Study 4: Main
experimental study’ along with its findings.
147
6.2. Method
6.2.1. Participants
80 university participants were recruited as an opportunity sample from UK (N = 43) and India (N
= 37) with an average age of M = 23.14 (SD = 5.02) and an age range between 18 to 41 years.
6.2.2. Procedure
The survey was conducted in paper format for all participants in the UK and in India. For the
Indian sample a translated version of the survey was used. The survey lasted for approximately
15-20 minutes. Full participant consent was attained before their participation in the survey and
were also given the chance to enter in a lottery draw to win gift voucher as a thank you for their
participation which was completely voluntary.
6.2.3. Material & Description of items
All the priming tasks and the measure for the effect of priming had been adapted from previous
validated research. All the tasks used had high effect sizes and has been used in inter-cultural
context before. As the priming tasks were short tasks therefore it was decided to combine all the
three tasks (The pronoun circling task, short writing task & Sumerian story task) in order to
enhance the priming effect. The demographic information of the participants was collected and
coded as below:
Gender coded as 1 for male and 2 for female, age in years, ethnicity coded as 1 for respondents
with western ethic and 2 for any eastern ethnicity, work coded as 1 for employed –full time, 2 for
employed – part time, 3 for unemployed, 4 for student – full time, 5 for student – part time, country
of residence, coded as 1 for UK and 2 for India, nationality coded as 1 for British and 2 for Indian
and first language of the participants which was coded as English as 1 and Bengali as 2.
This was followed by cultural self-construal measured by a scale adapted from (Singelis, 1994).
The scale consisted of 12 items measuring the collectivist self-construal (Cronbach’s alpha = .61)
148
and consisted of 12 items measuring the individualist self-construal (Cronbach’s alpha = .57) of
the participants. Item wordings in the scale were modified slightly to fit in with the online
environment and the research aims. For a full list of the items see Appendix 3. The scale was
measured on a 5 point Likert scale ranging from “Disagree strongly” coded as “1” to “Agree
strongly” coded as “5” where “1” was the least score and “5” was the highest score on the scale
which indicated levels of collectivist orientation for the collectivism self-construal scale and levels
of individualism for the individualism self-construal scale. The design of the experiment was
framed in such a way so that at first the questionnaire helped to establish the cultural orientation
of the participants and then participants were exposed to the priming tasks which hoped to
manipulate their cultural selves. Participants were asked to respond to further tasks which checked
whether their decisions they made after they were primed changed with the cultural orientation
they were primed with. For example, if participants scored higher in the level of collectivism prior
to priming, then they would score higher in their collectivist self-descriptions when primed with
their consistent cultural orientation then when they are primed with inconsistent cultural
orientation. Similar effect would be expected when participants are given individualism version as
participants who scored higher on individualism self-construal scale prior to priming, would score
higher in self-descriptions related to individualist attributes when primed with individualism
version of the questionnaire than when they are primed with the collectivist version of the
questionnaire. Participants were randomly given the two versions of the questionnaire. For the
Indian sample, the study was conducted with the researcher on site to help with any queries or
questions on the spot. A combined score was also calculated to check the scores for collectivist
and individualist self-construal before priming. The overall score for collectivism before priming
was computed by including all the items in the self-construal scale that measured interdependence
and the reversed scored items which measured the independent self-construal. Similarly, for
computing the overall score on individualist self-construal before priming, all the items in the self-
construal scale that measured independent self-construal was included and the reversed scored
items measuring the interdependence self-construal was taken. These two overall scores were both
standardized using a z-transformation.
The participants were subjected to a combined priming manipulation consisting of three
procedure established in the research literature. Tasks were selected based on their format, but also
on effect sizes reported in the overview provided by Oyserman and Lee (2008). First, participants
149
were asked to work on a pronoun circling task adapted from (Brewer & Gardner, 1996) and
Gardner et al. (1999, study 1). Documented effect sizes for this task ranged from d = 0.24, for
European Americans, to d = 0.84, for Hong Kong Chinese participants (see Oyserman & Lee,
2008). The task consisted of a text passage that described a trip to the park and respondents were
asked to read the passage and circle all the pronouns in the passage. The two conditions of the
pronoun circling task were created by setting up an independent version of the text where the
pronouns were “I, mine, me, and myself” and an interdependent version where the pronouns were
“we, they, us, and ourselves”. For collectivism priming the text ran: “We go to the park. Our
excitement fills us when we see the ice-cream van. We allow ourselves to explore every corner of
the park, never letting other people distract us. Our voice fills the air and street. We window shop
and everywhere we go we see our reflection looking back at us in the glass from the shops we walk
past. When we return home, our hearts fill with joy and happiness as we know that we will soon
return back in the park. The park belongs to us”. For individualism priming the text was: “I go to
the park. My excitement fills me when I see the ice-cream van. I allow myself to explore every
corner of the park, never letting other people distract me. My voice fills the air and street. I window
shop and everywhere I go I see my reflection looking back at me in the glass from the shops I walk
past. When I return home, my hearts fills with joy and happiness as I know that I will soon return
back in the park. The park belongs to me”.
The priming manipulation was further reinforced by additional tasks. Participants were next
primed with a short writing task adapted from (Trafimow, Triandis, & Goto, 1991) which had a
previously reported average effect size of d = 0.42 (conducted on participants in the U.S.; see
(Oyserman & Lee, 2008). Instructions for individualism priming were: “For the next two minutes
you will need to write 2 sentences based on these instructions below: a) Please think of what makes
you different from your family and friends?, b) What is your personal goal for the next 1 year?”.
For collectivism priming participants were instructed as follows: “a) Please think of what you have
in common with your family and friends?, b) What is your goal for the next 1 year in relation to
your family and friends?”. In order to increase participants’ engagement with the procedure they
were asked in both conditions to first think about the task and then to write down their response in
the survey. This differed from the original study conducted by (Trafimow et al., 1991) where
participants did not have to write anything, but had to just think about the task for two minutes.
150
The final reinforcement of the priming manipulation was achieved with two versions of the
‘Sumerian story’, an interdependence-focused or independence-focused story of a general who
was in a dilemma when he had to choose an appropriate and deserved warrior to send to his king.
This method was adapted from (Gardner et al., 1999; Trafimow et al., 1991) and had previously
shown an effect size of d = 0.88 (conducted on European Americans; see Oyserman & Lee, 2008).
The interdependent version of the story was: “Sostoras, a warrior in ancient Sumer, was largely
responsible for the success of Sargon I in conquering all of Mesopotamia. As a result, he was
rewarded with a small kingdom of his own to rule. About 10 years later, Sargon 1 was conscripting
warriors for a war. Sostoras was obligated to send a detachment of soldiers to aid Sargon 1. He
had to decide who to put in command of the detachment. After thinking about it for a long time,
Sostoras eventually decided on Tiglath who was a member of his family. This appointment had
several advantages. Sostoras was able to show his loyalty to his family. He was also able to cement
their loyalty to him. In addition, having Tiglath as the commander increased the power and prestige
of the family. The independent version of the story was: “Sostoras, a warrior in ancient Sumer,
was largely responsible for the success of Sargon 1 in conquering all of Mesopotamia. As a result,
he was rewarded with a small kingdom of his own to rule. About 10 years later, Sargon 1 was
conscripting warriors for a war. Sostoras was obligated to send a detachment of soldiers to aid
Sargon 1. He had to decide who to put in command of the detachment. After thinking about it for
a long time, Sostoras eventually decided on Tiglath who was a talented general. This appointment
had several advantages. Sostoras was able to make an excellent general indebted to him. This
would solidify Sostoras's hold on his own dominion. In addition, the very fact of having a general
such as Tiglath as his personal representative would greatly increase Sostoras's prestige. Finally,
sending his best general would be likely to make Sargon 1 grateful. Consequently, there was the
possibility of getting rewarded by Sargon 1”. In both the versions participants were then asked if
they admired Sostoras for his decision. These responses, in line with previous studies, were taken
as an indication that participants had engaged with the story and were not actually considered for
analysis.
This was then followed by the Twenty Statements Task (TST) adapted from (Gardner, Gabriel,
& Lee, 1999; Kuhn & McPartland, 1954) which had an average effect size d = 0.67, see (Oyserman
& Lee, 2008) and was the DV which helped to establish the effect of priming on participant
responses. Given the brevity of the experimental procedure, and the manipulation in particular, it
151
was decided not to repeat the cultural self-construal scale from the beginning. This would have
allowed for an easy comparison of values pre and post priming, but would have required
participants to fill in exactly the same measure within a few minutes. The TST was chosen as an
equivalent but alternative measure of self-construal. Participants were asked to give 20 self-
descriptions of themselves. Responses were then coded as independent, coded as “1”, if
respondents described a person attribute (trait, feeling, physical descriptor or attitude – e.g., “I am
selfish”) and as interdependent, coded as “2”, if they described a collective relationship or
membership with other group – e.g., “I am happy to be with my family”. A total score for each
participant was calculated and the overall scores were standardized using z-transformation.
Finally, participants were thanked for their participation and were debriefed. The below section
will now look at the results and analysis of this first experiment.
6.3. Results and Analyses
This section will discuss the results and the analysis that followed after the data collection which
will follow with a brief summary.
6.3.1. Data screening and preliminary analysis
Data was analysed using SPSS version 23. The threshold for significance was set to α = 0.05
for all decisions. Prior to in-depth analysis data were first screened to check for missing data and
outliers. Boxplots and subsequent checks revealed the presence of some outliers and missing data
which was coded as 99 in SPSS. The assumptions of normality were also checked through and
inspection of skewness or kurtosis for each variable. Normality of the data was accepted if the
distribution of a variable, i.e., both skewness and kurtosis, showed acceptable properties (z < 2.58)
(Hair et al., 2006).
Table 6.1. Below demonstrates the characteristics of the participants. The overall mean age of
the UK respondents was 21.97 years (SD = 3.24) with a range between 18 and 33 years. The mean
age of the Indian respondents was 22.75 years (SD = 6.17) with a range between 18 and 47 years
suggesting sufficient variation of age in the data, comparable across countries. Although a large
majority of participants came from a student population, the sample also showed participation from
working members (see Table 6.2).
152
Table 6.1.
Illustrates the Descriptive statistics for participants in Study 3
Variable Mean (SD)/ Frequency Range/ Percentage
Overall (N= 70, missing 1)
Age (years):
UK 21.97 (3.24) 18-33
India 22.75 (6.17) 18-47
Country:
UK: N = 34; India: N = 36, missing 1)
Gender:
UK : Male 12 35.3
: Female 22 64.7
India : Male 05 13.9
: Female 31 86.1
Note: Overall represents both UK and Indonesia combined
Age: Missing from Indian Sample N = 1
Table 6.2.
Ethnicity and work characteristics of the participants in Study 3
Variable Frequency Percentage
Overall (N= 71)
Ethnicity:
153
UK: British 34
India: Bengali 37
Work:
UK: (N = 34)
Employed (full time) 2 5.9
Employed (part time) 3 8.8
Unemployed - -
Student (part-time) - -
Student (full-time) 29 85.3
India: (N = 36; missing N = 1)
Employed (full-time) 2 5.6
Employed (part-time) 1 2.8
Unemployed - -
Student (part-time) 4 11.1
Student (full-time) 29 80.6
Note: Overall represents both UK and Indonesia combined
The next step was to understand the relationship between the variables in the study for which a
Pearson’s bivariate correlation was conducted using SPSS version 23 (see Table 6.3). Table 6.3.
confirms that there were some significant relationships between variables used in the study. For
example, increase in collectivism before priming decreased scores on individualism before
priming as there was a negative correlation between these two items (r = -.999, p < 0.01) as
expected. Furthermore, increase in age increased gender (r = .290, p < 0.05) suggests that as age
154
increase, there is an increase in female members on SNS. Country was coded as 1 for UK and 2
for India. The results below confirm that Indian participants had a higher score on the prime after
test which was the scores on TST scale (r = .345, p < .001), higher country score also increased
individualism score before priming (r = .364, p < 0.01), increase in collectivism score before
priming (r = .368, p < .001), which confirms that the Indian participants had higher scores on both
cultural traits. Increase in country also increased gender (r = .249, p < 0.05) which confirms that
there were a higher number of female participants in the Indian sample. The results clearly
highlighted that there existed a correlation between the country differences and among most of the
variables used in the study due to which it was decided that country along with priming be used as
fixed factor.
Table 6.3.
Demonstrates the Inter-Correlation of the variables in Study 3
Prime After IND Before COLL Before Prime Condition Gender Age Country
Prime After
1
IND Before -.039 1
COLL Before .047 -.999** 1
Prime Condition .181 -.168 .163 1
Gender .190 -.021 .017 -.046 1
Age -.013 -.174 -.178 .144 .290* 1
Country .345** .364** .368** .121 .249* .079 1
Note: Correlation is significant at the 0.05 level (2-tailed) *
155
Correlation is significant at the 0.01 level (2-tailed) **
Prime After = score after priming on TST scale, IND Before = score of individualism before priming,
COLL Before = score of collectivism before priming
N = 71
A 2 (country type: UK vs India) x 2 (prime type: independent vs interdependent) analyses of
covariance (ANCOVA) was used to analyse the effect of priming on the standardized TST (UK:
Mean = -.33, SD = .92; India: Mean = .32, SD = .88) scale. The mean score of self-construal scale
(collectivism before priming (UK: Mean = -.38, SD = 1.04; India: Mean = .34, SD = .80) and
individualism before priming (UK: Mean = .37, SD = 1.04; India: Mean = -.34, SD = .81) along
with age and gender was used as a covariate to eliminate the effect of any variance on the final
scores.
No significant main effect was found for priming conditions on the TST scores F(1,61) = 2.69,
p = .11, partial η2 = .04 which confirms that priming failed to show any effect on participant
responses. There was a significant main effect for country found on the TST scores F(1,161) = .15,
p = .03, partial η2 = .07 suggesting that participants in the UK scored less on the TST scale (M = -
.37) than participants in India (M = .21) and the difference in the two groups was significant p
=.03. However, there was no significant interaction found between priming conditions and country
F(1,61) = .15, p = .70, partial η2 = .00 which was reflected in the weak effect see Figure 6.1.
No significant main effect was found for collectivism before priming: F(1,61) = 1.89, p > 0.05,
partial η2 = .03, individualism before priming F(1,61) = 1.96, p > .05, partial η2 = .03, gender
F(1,61) = .72, p > .05, partial η2 = .01 and age F(1,61) = .14, p > .05, partial n2 = .00.
156
As priming failed to show any significant effect, further analysis was conducted where country
was not used as a covariate in a between subject analysis of covariance (ANCOVA). The results
confirm that the effect of priming condition did not change F(1,61) = 2.62, p = .11, partial η2 =
.04. This again failed to show any changes in the effect on priming.
6.4. Summary
Study 3 confirmed that there existed a relationship between variables, however while the main
aim of this study was to understand if the priming conditions would have an effect on participant
responses on the TST task, it could not be established. Even ruling out the effect of country on the
participant scores on the TST scale, the effect of the priming conditions could not be achieved. It
is however argued that this could be due to the sample size or the task itself. Therefore, this called
for further exploration in Study 4.
6.5 Changes made after completion of Study 3
While Study 3 failed to show any significant effect of priming conditions on participant
responses, it was important to evaluate if this was due to the study itself. Therefore, after
completion of the study, feedback from participants was collected and analysed to help develop
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
UK India
Mea
n r
atin
gs
Priming
Figure 6.1. Graphical illustration of the interaction effect of country and priming on TST scores in UK and India
Individualism Collectivism
157
Study 4. All the feedback was checked and amendments were made after considering how the
manipulation could be strengthened. The feedback received was quite similar to the ‘Study 1 –
online survey’. Some of the comments made were focused on issues of instruction and
communication (‘more clarity required’, ‘could not relate to Sastoras task’ (this was the Sumerian
warrior task) and ‘the names of the characters sounded similar’, ‘the twenty items tasks were too
long’). This was in line with some additional face-to-face feedback obtained during data collection.
On the basis of the feedback received, the instructions for the tasks were made clearer and
shorter, e.g., for the TST in Study 3 the instructions were: ‘There are twenty number blanks on the
page below. Please write twenty answers to the simple question ‘Who am I?’ in the blanks. Just
give twenty different answers to this question. Answer as if you were giving the answers to
yourself, not to somebody else. Write the answers in order that they occur to you. Don’t worry
about the logic or ‘importance’. Go along fairly fast as time is limited’. For the amended version
for Study 4 this was changed to: ‘There are ten number blanks on the page below. Please write ten
answers to the simple question 'Who am I?'. Just give ten different answers to this question. Answer
as if you were giving the answers to yourself, not to somebody else. Write the answers in order
that they occur to you. Don't worry about the logic or 'importance'. Go along fairly fast as time is
limited. For example: I am "a student", I am "a daughter"., etc.
In order to better capture cultural orientation of the participants, the measures were reconsidered
and a one item scale was also introduced which highlighted cultural traits and behaviours (see
Markus & Kitayama, 1991). Additionally, the name of the king in the ‘Sumerian warrior story’
was changed from ‘Sargon 1’ to ‘Gilgamesh’. Furthermore, in the ‘Pronoun circling task’,
participants were also asked to in addition to circling the pronouns they were also required to count
the number of pronouns. This was done to enhance participant engagement with the task. Also, in
the Twenty Item Test (TST), the number of self-descriptions was reduced from 20 to 10. An
additional self enhancement scale consisting of 7 collectivism measuring attitudes and 7
individualism measuring attitudes were included in the questionnaire which was measuring before
and after the priming tasks. Other additional scales that were included were ‘Online Trust scale’,
‘Online Privacy scale’, ‘Online group identification scale’, ‘Online Social Support scale’ & several
scenarios (which will be discussed in detail in the Chapter 7).
158
Additionally, while it was seen that the scale reliability of some of the items like the cultural
self-construal scale only achieved a medium scale reliability it was decided that all the 24 items
would now be included in the study as the reduced version might have affected the scale reliability
score. It was also decided that the individual self-construal items and the collectivism measuring
items would be kept as separate individual scales. Additional to this the items in the self
enhancement scale which measured the collectivist self attributes had a moderate scale reliability
(Cronbach alpha = .56), however for the items that measured the individualism measuring
attributes was high (Cronbach alpha = .70). It could be argued that the difference in the values
could be due to the small sample size, hence it was decided to be used in the final survey as the
items had been validated in past research (Gaertner et al., 2008). The attributes were further
validated in the pilot study conducted by Sedikides et al. (2003). After the amendments were
confirmed the final version of the survey was developed on Bristol Online Survey (BOS) website
and distributed in UK and in Indonesia. Chapter 7 will discuss Study 4 in further detail.
159
Chapter Seven
Study 4: Experiment
(Priming and Facebook responses)
7.1. Rationale for this study
Study 3 helped to evaluate the impact of priming on participant responses. While it was a shorter
questionnaire, it did include the pre and post measures which helped to strengthen the decision to
use the priming measures in Study 4. The results in Study 6 highlight a near significant effect of
priming on participant responses which will be further evaluated in this study. As highlighted in
the last chapter various factors were identified that could have stood in the way of stronger effects
ranging from instructions in the paper-and-pencil based experiment to sample size. An amended
version of the materials was therefore used to conduct study 4, for which a larger sample was
recruited across Indonesia and UK. While study 1 and 2 looked at the differences in responses in
the measures, Study 3 and 4 aimed to evaluate the causal role of culture, through the priming
measures. Similar to Study 3, Study 4 consisted of the priming measures but it also had additional
measures which evaluated the effect of priming on several DVs (which will be discussed later).
Next to measures of self-construal, the same main outcome measures as in the surveys were added
to the experiment: online self-enhancement, online privacy, online trust, online group identity and
perceived social support. This study aims to evaluate the following hypotheses:
H12: Participants will respond more strongly when they are primed with their consistent
cultural self then when they are primed with inconsistent cultural self, such that members
from collectivist countries will score higher on levels of collectivism when they are primed
with collectivism self-construal than when they are primed with individualism self-construal.
Similarly, when members from individualist countries are primed with individualism self-
construal they will score higher on the level of individualism than levels of collectivism.
It is envisaged that when members from particular cultural background are primed with their
own cultural attributes, it will result in enhancing their cultural selves even deeper, thereby helping
to strengthen their cultural orientation. Alternatively, when they are exposed to an environment
160
which promotes cultural attributes that are not consistent to their own than that would also have
an impact on their cultural selves.
H13: Priming would have a positive effect on group identification when primed with
consistent cultural orientation than when primed with inconsistent cultural orientation.
Furthermore, Indonesian sample would score higher on group identification than UK
sample.
While the social network on SNS extends and gets established from links that members know
offline (Lampe et al., 2006), it would be fair to say that members from both countries will have
social contacts that they already know offline, however as members from collectivist countries
prefer staying within their social group, their social network would be less diverse than members
from individualist countries and as such they would be more familiar with their social network.
H14: Priming would have an effect on how members rated themselves when compared with
others on cultural attributes, such that it helps them to self-enhance. It is further anticipated
that members from both cultures would rate themselves higher on their cultural traits.
As SNS becomes the environment where members are able to advertise themselves through
their online identity, it is envisaged that social comparison with other members online on cultural
specific attributes would help to highlight how members would perceive themselves in comparison
to others. When members are primed with their cultural specific attributes it will strengthen their
cultural attributes which will help them to self-enhance. Alternatively, when members are primed
with inconsistent cultural attributes it will lower their feeling of self-enhancement.
H15: Priming would have an effect on social support when members are primed with
consistent cultural orientation than when they are primed with inconsistent cultural
orientation.
It is anticipated cultural orientation of members would have an effect on how members are able
to perceive social support in an online environment. It is envisaged that consistent cultural traits
among members would help them seek social support more than when they are communicating
with members with inconsistent cultural traits. It is because as members are aware of cultural
specific behaviour it helps them anticipate the online behaviour of their social contacts. Such
161
predictions might not be easy to make when they communicate with members from other cultural
backgrounds, who have different cultural traits.
H16: Priming would have an effect on online trust. It is hypothesized that when members
are primed with collectivism condition then it would increase online trust as it is anticipated
that Indonesian participants would score higher in level of online trust than the UK
participants. Alternatively, when members are primed with individualism condition then it
would reduce their level of online trust for their social contacts.
When members are primed with consistent cultural orientation, it will have an effect on the
levels of trust on SNS. It is anticipated that when members from collectivist countries are primed
with collectivism condition, it will strengthen their cultural selves further which will have an effect
on the level of trust among their social network. It is anticipated that as members from collectivist
countries would trust their social network more on Facebook more than members from
individualist countries as their social network would generally consist of members with whom they
would have established offline relationships unlike members from individualist countries who are
open to accepting friend requests from diverse social groups. Furthermore, collectivist attributes
would promote interdependent selves of member which would help to establish a feeling of
reciprocity among in-groups which would instil a feeling of trust among members.
H17: Priming would have an effect on levels of online privacy concern such that members
from collectivist countries would score higher on the scores on privacy concern than
members from individualist countries.
It is anticipated that as members from collectivist countries prefer to remain within their own
known social groups they would be more concerned about maintaining their privacy in an online
environment than members from individualist countries. It is because failure to maintain
appropriate privacy could result in online tension which could also have an adverse impact on their
offline relationships.
162
7.2. Method
7.2.1. Participants
407 participants were recruited as an opportunity sample from a university in UK (N = 161)
and across three universities in Indonesia (N = 246) with an average age of M = 19.57 (SD = 3.04)
and an age range between 18 to 44 years. The participants in Indonesia were recruited with the
help of a Research Assistant.
7.2.2. Procedure
The experiment was developed by creating different versions of an online survey, similar to
study 3. In the UK, the online versions were promoted within the host university and on Twitter
and Facebook to maximize outreach. For the Indonesian sample, the survey was first back
translated and the data was collected in paper and pen format and then was inputted on the Bristol
Online Survey (BOS) website by research assistants recruited for this task. The experiment lasted
for approximately 15-20 minutes. A small incentive was offered as participants were given the
chance to enter in a lottery draw to win various prizes for the UK sample and vouchers for the
Indonesian sample.
7.2.3. Materials &Description of items
The experiment captured the demographics of the participants. Gender coded as 1 for male and
2 for female, age in years, ethnicity coded as 1 for respondents with western ethnicity, who were
White British and 2 for Eastern ethnic background who were Javanese, work coded as 1 for
employed –full time, 2 for employed – part time, 3 for unemployed, 4 for student – full time, 5 for
student – part time, country of residence, coded as 1 for UK and 2 for Indonesia, nationality coded
as 1 for British and 2 for Indonesian and first language of the participants which was coded as
English as 1 and Bahasa Indonesia as 2.
While the DV in Study 3 consisted of the responses on the TST scale, Study 4 consisted of
several additional DVs which were all standardized measures. The experiment began with a check
on the demographics of the participants. The cultural orientation of the participants was captured
163
using the cultural self-construal scale (Singelis, 1994) which, as in the other studies, measured the
collectivist and individualist self-construal of the participants. All the 24 items were used (12 items
measuring the levels of collectivism and 12 items measuring the levels of individualism).
Respondents were asked to give their best rating on the items based on a 5 point Likert scale
measuring from “Disagree strongly” coded as “1” to “Agree strongly” coded as “5” where “1”
being the lowest score and “5” being the highest score. The collectivism-measuring items were
given first followed by the individualism-measuring items (Cronbach’s alpha = .75 for
individualism; Cronbach’s alpha = .67 for collectivism).
In light of moderate reliability levels for the self-construal scale, found across all studies of the
project, an additional two item scale which was designed and adapted from the cultural
descriptions from Markus & Kitayama (1991). The single items were designed in a way which
promoted general description of cultural traits and behaviours. ‘I value individual achievement and
strongly believe in my personal goals. My personal achievements and success is of priority for me.
I strongly believe in my own efforts. I value independence’, highlighted levels of individualism
and ‘I value the importance of my relationship with others. My goals are achievable/ achieved with
the help of and support of others around me. Living a life of harmony with others is of priority for
me’ highlighted collectivist behaviours. Both items were also based on 5-point Likert scales
ranging from ‘Not at all’ to ‘Very much’ where ‘Not at all’ was coded as ‘0’ and ‘Very much’ as
‘5’.
In order to evaluate the effect of priming on the levels of self-enhancement on cultural specific
traits, a self-enhancement measure was adapted from Gaertner et al. (2008) and was administered
twice, before and after the manipulation. The scale consisted of 7 collectivist attributes and 7
individualist attributes. Respondents were asked to rate the attributes based on the importance of
the traits to them “personally”. Ratings were made on a 6 point Likert scale ranging from “very
unimportant” coded as “1” to “very important” coded as “6” (Cronbach’s alpha = .73). Higher
scores indicated higher levels of cultural traits.
‘The Sumerian warrior task’ was used as one of the priming instruments. As in study 3, the
participants were primed using an interdependent or independent story of a general who was in a
dilemma when he had to choose the appropriate and deserved warrior to send to his king (Gardner
et al., 1999; Trafimow, Triandis, & Goto, 1991). The two stories aim to bring about a shift between
164
independent and interdependent self-construal. In the independent condition of the warrior story,
the general chose the warrior that was best suited for the job and considered the benefits to himself.
Whereas in the interdependent version of the story the general chose a member of his family and
considered the benefits to his family. Respondents were asked to read the story and, in addition, to
decide if they thought the general’s decision was supportive. The next priming task was the
‘Pronoun circling task’ adapted from Brewer & Gardner (1996). This text passage described a trip
to the park and participants were asked to read the passage and circle all the pronouns in the
passage. On addition to this task, in comparison to study 3, participants were asked to circle the
pronouns present in the passage and also write down the number of pronouns they found. This was
done to enhance participant engagement. Two conditions of the pronoun circling task were created
with an independent version of the text where the pronouns were “I, mine, me, myself” and an
interdependent version where the pronouns were “we, they, us, ourselves”. In total, there were
seventeen pronouns in the task in both the versions. The third priming task required participants
to write down answers to four questions. For the collectivist version, questions were ‘a) List 4
things that you would like to obtain for your family to improve their everyday life, b) List 4 things
that you value about your family, c) Please think of 4 things that you have in common with your
family and friends and d) What is your goal for the next 1 year in relation to your family and
friends, i.e., what is it that you want to achieve or do together with your family and friends in the
next 1 year?’. For the individualist version, the questions were ‘a) List 4 things that you would like
to obtain for yourself to improve your everyday life, b) List 4 things that you value about yourself
as a person, c) Please think of what makes you different from your family and friends and d) What
is your personal goal for the next 1 year.’ All the priming tasks were presented to participants one
after the other to strengthen the effect of priming on the participants.
Respondents then completed a self-construal task or TST (Kuhn & McPartland, 1954), which
the DV. The original version of the TST required respondents to provide twenty self-descriptions.
However, based on the responses and the feedback received in Study 3 it was seen that reducing
the number of self-descriptions would be less time consuming and would encourage respondent
engagement with the task. Therefore, instead of twenty self-descriptions respondents were asked
to provide ten self-descriptions. Responses were then coded as “independent” and scored as “1” if
respondents described a person attribute (trait, feeling, physical descriptor or attitude – e.g., “I am
165
selfish”) and as “interdependent” and scored as “2” if they described a collective relationship or
membership with other group – e.g., “I am happy to be with my family”.
The online privacy scale consisted of items like “How confident are you that the information
that you upload on Facebook is not misused by others?”, “How safe do you feel uploading your
pictures and personal information on Facebook?”, “While on Facebook do you feel that you are at
a particular risk?” and “Do you feel that others are at risk because of using Facebook?”. The items
were measured on a five point Likert scale ranging from “Not at all” coded as “1” to “Very much”
coded as “5” (Cronbach’s alpha = .65).
The experiment included additional items which were items that focused on some of the social
drivers that drives people on SNS which helped to highlight some of the motives of online
behaviour in more detail. The items in the scale were: “I feel confident when my friends appreciate
my achievements on Facebook”, “Sharing harmony among my social groups on Facebook is
crucial to me”, “Sharing my personal photos and information gives me a sense of freedom”, I feel
valued and appreciated when my friends share their likes and comments on my personal photos
and information on Facebook”, and “My popularity on Facebook depends on the number of friends
I have on Facebook”. Items were aggregate to form a scale called the “Social Drivers Scale” with
Cronbach’s alpha = .73.
The social support scale consisted of four items “How likely would you offer social support to
others”, “How likely would you share your most private worries and fears with someone in your
social group on Facebook?”, “How likely do you feel that members of your social group would
help you during a personal crisis” and “How likely are you to involve yourself with a social cause
on Facebook?”. The items were rated on a 5 point Likert scale ranging from “Extremely unlikely”
coded as “1” and “Extremely likely” coded as “5” (Cronbach’s alpha = .54).
The group identity scale consisted of four items “I feel strongly connected with my social group
members on Facebook”, “My Facebook friends are very important for me”, which were positively
rated items. “I often feel held back by my social group on Facebook”, “I sometimes make excuses
of belonging to my friends group on Facebook” were negatively rated items. The negatively rated
items were recoded. The items were rated on a 5 point Likert scale ranging from “Never” coded
as “1” to “Always” coded as “5”. The Cronbach’s alpha for this scale was = .64.
166
The online trust the members had for their social contacts on Facebook was measured using a
scale consisting of four items: “My Facebook social group members are trustworthy”, “I am
confident that my friends on Facebook mostly share their honest opinions about me”, “I trust
Facebook for safeguarding my personal information”, and “Members of my social group on
Facebook do their best to help me whenever I need them”. The items were coded on a 5 point
Likert scale ranging from “Never” coded as “1” to “Very Often” coded as “5”. Cronbach’s alpha
for this scale was = .77.
The respondents were then presented with some scenarios typical of Facebook engagement: “I
log on my Facebook and I receive a friend request from my family member. How likely is that you
will accept the friend request?”, “I log on my Facebook and receive a friend request from an
unknown person. How likely would you accept the friend request?”, “While logging on Facebook
I am reminded to update my privacy settings. How likely would you update your privacy
settings?”, “While on Facebook I get a request to join a protest for a noble/ social cause. How
likely would you join the protest?”, “While on Facebook when I see my friends talk about their
achievement, I would congratulate them”, “While on Facebook I often seem to join social groups
that has personal importance to me”. Item responses could range from “Extremely unlikely” coded
as “1” to “Extremely likely” coded as “5”.
There was also an overall check built into the experiment to ensure that participants had some
basic understanding and awareness of the priming tasks. For this, participants were asked to focus
back on the study as a whole and to state what they thought was the study’s main focus: “Social
responsibility”, “Individual decisions” or “Reading skills”. The three items were scored on a 7
point Likert scale as well ranging from “Not at all” coded as “1” to “Very much” coded as “7”.
7.3. Results and Analysis
7.3.1. Data screening and preliminary analysis
Data was analysed using SPSS version 23. The threshold for significance was set to α = 0.05
for all decisions. Prior to in depth analysis data were first screened to check for missing values and
outliers. The assumptions of normality were also checked if there was any presence of skewness
167
or kurtosis in the data. Normality of the data was accepted if the distribution of variables, i.e., both
their skewness and kurtosis was within range (z < 2.58), (Hair et al., 2006).
7.3.2. Sample characteristics
Table 7.1 demonstrates the characteristics of the participants. Mean age of the UK respondents
was M = 19.57 (SD = 3.04), with an age range between 18 and 44 years. The mean age of the
Indonesian respondents was M = 19.18 (SD = 1.30) years. The age range for the Indonesian
respondents was between 18 to 30 years. The age range for the UK respondents was more varied
than the Indonesian sample, however overall the sample was clearly university-based in both
countries, with more than 90 percent of participants coming from a student population. This is also
demonstrated in Table 7.2 which shows the variation in work status for both countries. The highest
number of participants were from the population from 18 – 21 years of range. The sample also
showed a greater number of participation from the female population in both UK and Indonesia.
Table 7.1.
Illustrates the Descriptive statistics for participants in Study 4
Variable Mean (SD)/ Frequency Range/ Percentage
Overall (N= 407)
Age (years):
UK 19.57 (3.04) 18-44
Indonesia 19.18 (1.30) 18-30
Country: UK: N = 161; Indonesia: N = 246)
168
Gender:
UK : Male 18 11.2
: Female 143 88.8
Indonesia : Male 52 21.1
: Female 194 78.9
Note: Overall represents both UK and Indonesia combined
Table 7.2.
Ethnicity and work characteristics of the participants in Study 4
Variable Frequency Percentage
Overall (N= 407)
Ethnicity:
UK: British 161
Indonesia: Indonesian 246
Work:
UK: (N = 161)
Employed (full-time) 0 0.0
Employed (part-time) 4 2.5
Unemployed 4 2.5
Student (part-time) 2 1.2
169
Student (full-time) 151 93.8
Indonesia: (N = 246)
Employed (full-time) 1 .4
Employed (part-time) - -
Unemployed - -
Student (part-time) 5 2.0
Student (full-time) 240 97.6
After checking for the demographics of the respondents in the UK and in Indonesia a series of
2 (country type: UK vs Indonesia) x 2 (prime type: independent vs interdependent) ANCOVAs
was conducted on all the outcome variables. Age and gender were used as covariates in all the
analyses in order to make sure that any effect due to the difference is controlled for. The below
paragraphs will discuss the findings.
7.3.3 Twenty Statements Test (TST) of Self-construal
The following hypothesis were evaluated:
H12: Participants will respond more strongly when they are primed with their consistent
cultural self then when they are primed with inconsistent cultural self, such that members
from collectivist countries will score higher on levels of collectivism when they are primed
with collectivism self-construal than when they are primed with individualism self-construal.
Similarly, when members from individualist countries are primed with individualism self-
construal they will score higher on the level of individualism than levels of collectivism.
A 2 (country type: UK vs Indonesia) x 2 (prime type: independent vs interdependent) analysis
of covariance (ANCOVA) was used to analyse the effect of priming on the standardized TST scale.
The cultural orientation scores prior to priming was used as a covariate to eliminate the effect of
170
any variance of cultural orientation on the scores after priming. This analyses of priming effect
showed that individuals had an effect on their self-construal when primed with a particular cultural
trait.
There was a significant main effect was found for country F(1,396) = 100.913, p < 0.05, partial
η2 = .20, where participants in the UK scored lower (M = -.47) than the Indonesian participants (M
=.45). This is in line with the coding in the task as higher scores related to collectivist self-
descriptions and were allocated “2” and lower scores of “1” were allocated to individualist self-
descriptions. This was as expected as participants in the UK which represented independent self-
construal would generally provide individualist self-descriptions of themselves than Indonesian
participants. It was expected that the Indonesian participants would generally provide a higher
number of interdependent self-descriptions of themselves as they represent a collectivist country.
There was a significant main effect for priming F(1,396) = 41.18, p < 0.05, partial η2 = .09,
where participants who were primed with individualism self-construal scored lower on the TST
scale (M = -.30) than when they were primed with collectivism self-construal (N = .27) This was
also further supported by the significant interaction between country x priming F(1,336) = 29.62,
p < 0.05, partial η2 = .07 on the scores on the TST. The results further confirm that when primed
with individualism self-construal participants in the UK (M = -.52) scored lower than the
Indonesian participants (M = -.07). Furthermore, when participants were primed with collectivism
self-construal, the Indonesian participants scored higher (M = .98) than the UK participants (M =
-.43). This is again as expected as when Indonesian participants, who generally have higher levels
of collectivist traits in them would have an increase in their feeling of collectivism or
interdependence when primed with their consistent cultural self than when they are primed with
inconsistent cultural self, i.e., when primed with collectivism self-construal than when they are
primed with individualism self-construal. This is also the case for the UK participants, however as
they generally have higher levels of individualist traits, when primed even if they score higher on
collectivism than before, they would still be scoring less than the Indonesian participants as they
would have a higher feeling of collectivism (see Figure 7.1 below). This finding helped to support
H12 as the results confirmed that as predicted members scored higher when they were primed with
their consistent cultural attributes than when they are primed with inconsistent cultural attributes.
171
Covariates did not show significant associations with TST scores (for cultural orientation:
F(1,396) = .11, p > 0.05, partial η2 = .00; for gender: F(1,396) = 1.03, p > 0.05, partial η2 = .00; for
age: F(1,396) = 1.70, p > 0.05, partial η2 = .00) suggesting that the impact of one’s cultural self,
difference in age and gender did not have an impact on the responses on the TST scale.
Overall, the analysis of TST scores shows that the manipulation was successful in influencing
self-construal, although the effects seems to be weaker for UK participants than for Indonesian
participants.
7.3.4. Online Group Identity
A 2 (country type: UK vs Indonesia) x 2 (prime type: independent vs interdependent)
ANCOVA was conducted on group identity with and age and gender used as covariates. Below is
the hypothesis that is evaluated:
H13: Priming would have a positive effect on group identification when primed with
consistent cultural orientation than when primed with inconsistent cultural orientation.
-1
-0.5
0
0.5
1
1.5
Individualism Collectivism
Mea
n s
core
s o
n T
ST s
cale
Priming
Figure 7.1.Graphical interpretation of the interaction effect on the TST responses
UK Indonesia
172
Furthermore, Indonesian sample would score higher on group identification than UK
sample.
No significant main effect was found for priming F(1,401) = 2.17, p > 0.05, partial η2 = .01
which failed to support H13. There was also no significant interaction effect between country x
priming interaction F(1,401) = .278, p > 0.05, partial η2 = .00, see Figure 7.2 which confirmed that
the combined effect of cultural orientation and one’s geographical residence did not have any
influence on how much members identified with their social network.
A significant main effect was found for country: F(1,401) = 18.70, p < 0.05, partial η2 = .05.
Participants in the UK showed higher levels of online group identity (M = .25) than participants
in Indonesia (M = -.19), see Figure 7.2. This further suggests that members in the UK can relate
to their social network on SNS more than the Indonesian participants.
For the covariates, no significant effect emerged (for gender: F(1,401) = .352, p > 0.05, partial
η2 = .00;, for age: F(1,401) = 1.78, p > 0.05, partial η2 = .00).
0.3
0.2
-0.09
-0.3-0.4
-0.3
-0.2
-0.1
0
0.1
0.2
0.3
0.4
Individualism Collectivism
Mea
n s
core
s o
n G
rou
p Id
enti
ty s
cale
Priming
Figure 7.2.Graphical interpretation of the interaction effect on Group Identity responses
UK Indonesia
173
7.3.5. Online Self-Enhancement
A 2 (country type: UK vs Indonesia) x 2 (prime type: independent vs interdependent) mixed
factor analysis of variance (ANOVA) was used to analyse the effect of priming on the self
enhancement scale, where self-enhancement was used as a within factor and country and priming
as between factors Age and gender were used as covariates. The below hypothesis was evaluated
through this analysis:
H14: Priming would have an effect on how members rated themselves when compared with
others on cultural attributes, such that it helps them to self-enhance. It is further anticipated
that members from both cultures would rate themselves higher on their cultural traits.
For the collectivist attributes there was no significant effect of priming on the ratings between
UK and Indonesia F(1, 401) = .334, p > 0.05, partial η2 = .001 suggesting that the causal role of
culture could not be established as such failed to support H14. There was also no significant
interaction effect established between country x priming F(1, 141) = 1.27, p > 0.08, partial η2 =
.003. However there was a significant effect of country difference on the ratings of on collectivist
attributes F(1, 141) = 6.46, p < 0.011, partial η2 = .016 which confirms that ratings on collectivist
attributes is affected by country differences such that participants in the UK scored themselves
lower (M = -.147) on collectivist attributes than Indonesian participants (M =.104). This was as
expected that members from Indonesia would advocate attributes that supports collectivist
behaviour and members from UK would advocate attributes that supported individualist
behaviour. See Figure 7.3 which confirms the above findings.
Age F(1, 141) = 1.165, p > 0.05, partial η2 = .003 and gender F(1, 141) = .194, p > 0.05, partial
η2 = .00 which were used as covariates failed to show any significant effect on the ratings of self-
enhancement on collectivist attributes.
174
For the individualist attributes there was no significant effect of priming on the ratings
between UK and Indonesia F(1, 401) = .019, p > 0.05, partial η2 = .00 suggesting that the causal
role of culture could not be established. There was also no significant interaction effect established
between country x priming F(1, 141) = .704, p > 0.05, partial η2 = .002. However there was a
significant effect of country difference on the ratings of on individualist attributes F(1, 141) =
20.104, p < 0.001, partial η2 = .048 which confirms that ratings on individualist attributes was
affected by country differences such that participants in the UK scored lower on individualist
attributes (M = -.26) than collectivist attributes (M = .18), see Figure 7.4.
Age F(1, 141) = .725, p > 0.05, partial η2 = .00 and gender F(1, 141) = .587, p > 0.05, partial
η2 = .001 which were used as covariates failed to show any significant effect on the ratings of self-
enhancement on individualist attributes.
-0.4
-0.3
-0.2
-0.1
0
0.1
0.2
pre post pre post
UK Indonesia
Mea
n s
core
on
sel
f-en
han
cem
ent
rati
ngs
Priming
Figure 7.3.Graphical interpretation of the interaction effect of pre and post priming ratings on Self-Enhancement on Collectivism attributes
Individualism Collectivism
175
7.3.6. Perceived Online Social Support
A 2 (country type: UK vs Indonesia) x 2 (prime type: independent vs interdependent)
ANCOVA was conducted on social support and age and gender were included as covariates. The
below hypothesis was evaluated through this analysis:
H15: Priming would have an effect on social support when members are primed with
consistent cultural orientation than when they are primed with inconsistent cultural
orientation.
There was no significant effect for priming on perceived online social support F(1,401) = 3.50,
p > 0.05, partial η2 = .01 which fails to support H14. There was also no significant effect found for
country (F(1,401) = 1.78, p > 0.05, partial η2 = .00) or the interaction (F(1,401) = .84, p > 0.05,
partial η2 = .00). No significant effects were found, neither for the covariates (for gender: F(1,401)
= 1.12, p > 0.05, partial η2 = .00; for age: F(1,401) = .012, p > 0.05, partial η2 = 00,).
-0.21-0.24
0.13 0.14
-0.26-0.32
0.210.25
-0.4
-0.3
-0.2
-0.1
0
0.1
0.2
0.3
pre post pre post
UK Indonesia
Mea
n s
core
on
sel
f-en
han
cem
ent
rati
ngs
Priming
Figure 7.4.Graphical interpretation of the interaction effect of pre and post priming ratings on Self-Enhancement on Individualist attributes
Individualism Collectivism
176
7.3.7. Online Trust
A 2 (country type: UK vs Indonesia) x 2 (prime type: independent vs interdependent)
ANCOVA was conducted on trust and age and gender were used as covariates with an aim to
evaluate the below hypothesis:
H16: Priming would have an effect on online trust. It is hypothesized that when members
are primed with collectivism condition then it would increase online trust as it is anticipated
that Indonesian participants would score higher in level of online trust than the UK
participants. Alternatively, when members are primed with individualism condition then it
would reduce their level of online trust for their social contacts.
The results confirmed that there was a significant effect of priming on trust scores between UK
and Indonesian participants F(1,401) = 8.626, p = .004, partial η2 = .021. When participants were
primed with individualism self-construal, they scored higher on online trust (M = .19) than when
primed with collectivism self-construal (M = -.10). No interaction between country and priming
was found: F(1,401) = .17, p > 0.05, partial η2 = .00. This confirms that the difference levels of
trust is effected by priming however the difference in scores was not as anticipated, hence failing
to support H16. There was also a significant effect of country difference. Participants in the UK
(M = 3.72) scored higher on the levels of trust with their social contacts on Facebook than
participants in Indonesia (M = -.276). F(1,401) = 43.011, p = .001, partial η2 = .097 further
confirming that country differences also influenced levels of trust among members, see Figure 7.5
below which explains the above.
No significant effects were found for covariates (for gender: F(1,401) = 2.80, p > 0.05, partial
η2 = .01; for age: F(1,401) = .04, p > 0.05, partial η2 = .00).
177
7.3.8. Online Privacy concern
A 2 (country type: UK vs Indonesia) x 2 (prime type: independent vs interdependent)
ANCOVA was conducted on the privacy scale and age and gender were used as covariates, and
the below hypothesis was evaluated:
H17: Priming would have an effect on levels of online privacy concern such that members
from collectivist countries would score higher on the scores on privacy concern than
members from individualist countries.
There was also a significant main effect of priming on the levels of privacy concern F(1,401) =
7.15, p < 0.05, partial η2 = .018, which supported the effect of priming on participant responses,
however priming participants with individualism attributes increases the level of privacy concern
(M = .214) than when they were primed with collectivism attributes (-.03), and the difference in
the two groups was statistically significant (p > 0.08) and hence failed to support H17. This could
be attributed to the level of education of the privacy risks associated with online communication
in different countries as members in UK might be better aware of the potential risks associated
with SNS, than members in Indonesia.
There was a significant main effect for country on the levels of privacy concern F(1,4010 =
119.31, p < 0.05, partial η2 = .23, where participants in the UK seemed to be more concerned about
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
UK Indonesia
Figure 7.5. Graphical illustration of the interaction effect of country and priming on the scores on Online Trust in UK and Indonesia
Individualism Collectivism
178
their privacy online (M = .57) than the Indonesian participants (M = -.40) and the difference
between the countries was statistically significant at p > 0.001 further suggesting that difference
in country residence had an impact on how members perceived online privacy. However there was
no significant interaction: F(1,401) = .869, p > 0.05, see Figure 7.6.
A significant effect of age on the levels of privacy concern was found F(1,401) = 4.57, p < 0.05,
partial η2 = .011 and there was a significant effect for gender on the levels of privacy concern
F(1,401) = 4.33, p < 0.05, partial η2 = 0.11 .
7.3.9. Scenario Task
A 2 (country type: UK vs Indonesia) x 2 (prime type: independent vs interdependent) between
subjects ANOVA design was used. Country and Prime type were the IVs and the score on the item
was the DV. The below paragraphs will show the results for individual items.
7.3.9.1. I log on my Facebook and receive a friend request from my family member. How
likely is that you will accept the friend request?
There was no significant main effect for priming F (1,400) = .555 p > .05, partial eta = .001, no
significant main effect for country F (1,400) = .828 p > .05, partial eta = .002, no significant main
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
0.8
UK Indonesia
Figure 7.6. Graphical illustration of the interaction effect of country and priming on the scores on Online Privacy Concern in UK and Indonesia
Individualism Collectivism
179
effect for age F (1,400) = .190 p > .05, partial eta = .000, no significant main effect for gender F
(1,400) = .00 p > .05, partial eta = .000, for cultural orientation F (1,400) = .767 p > .05, partial eta
= .002 and there was no significant interaction between priming and country F (1,403) = 1.301 p
> .05, partial eta = .003.
This suggests that accepting friend request on Facebook from family members did not affect
due to difference in country, cultural orientation, priming, age or gender.
7.3.9.2. I log on my Facebook and receive a friend request from an unknown person. How
likely will you accept the friend request?
There was a significant main effect for priming F (1,400) = 5.543 p < .05, partial n2 = .014,
significant main effect for country F (1,400) = .89.087, p < .001, partial n2 = .182. There was also
a significant interaction between country and priming F (1,400) = 5.619 p < .05, partial n2 = .014
which confirms that when members from UK were primed with individualism they scored the
same (M = -.57) when they were primed with collectivism self-construal (M = -.57). However,
when Indonesian participants were primed with individualism self-construal they would be more
likely to accept a friend request from an unknown person (M = .53) than when they were primed
with collectivism self-construal (M = .10). The difference in mean was quite small (.212) but was
statistically significant p = .02, see Figure 7.7 below. The results suggest that when having
collectivist attributes helps to avoid friending strangers on SNS.
There was a significant main effect age F (1,400) = 5.051 p < .05, partial n2 =. 012. However,
there was no significant main effect for gender F (1,400) = 3.627, p > .05, partial n2 = .009 and
differences in cultural orientation F (1,400) = .374, p > .05, partial n2 = .001.
180
7.3.9.3. While logging on Facebook I am reminded to update my privacy settings. How likely
would you update your privacy settings?
There was a significant main effect for priming F (1,400) = 6.443 p < .05, partial n2 = .016, a
significant main effect for country F (1,400) = 10.214 p > .01, partial n2 = .025 further confirming
that priming participants with individualist self-construal reduces the likelihood of the chances for
updating privacy settings when reminded (UK: M = -.31; Indonesia: M = .02)) than when primed
with collectivist self-construal (UK: -.06; Indonesia: M = .30) and the difference in mean in the
two groups was statistically significant p = .012. This suggests that the more individualist one is
the lesser they would update privacy settings. Alternatively, increase in collectivism would
increase the chances of updating the privacy settings on Facebook and such differences in
behaviour can be attributed due to the cultural differences. Adjusted means for country differences
suggest that participants in the UK seem to be less likely to update privacy settings when reminded
(M = -.18) than participants in Indonesia (M = .16) reflecting the previous findings, see Figure 7.8
below which illustrates these findings. There was also no significant interaction for country and
priming F (1,400) = .031, p > .05, partial eta = .000.
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
UK IndonesiaMea
n r
atin
gs
Priming
Figure 7.7. Graphical illustration of the interaction effect of country and priming (Scenario 2)
Individualism Collectivism
181
There was no significant main effect cultural orientation F (1,400) = .002 p > .05, partial n2 =
.00, there was also no significant main effect gender F (1,400) = 2.546 p < .05, partial n2 = .006,
there was also no significant main effect for age F (1,400) = 3.326 p > .05, partial eta = .008.
7.3.9.4. While on Facebook I get a request to join a protest for a noble/ social cause. How
likely would you join the protest?
There was no significant main effect for priming F (1,400) = .286 p > .05, partial η2 = .001 and
there was also no significant interaction between priming and country F (1,400) = .452, p > .05,
partial η2 = .001 which failed to support the causal role of culture on the differences in responses
among participants in UK and Indonesia. There was a significant main effect for cultural
orientation F (1,400) = 6.260 p < .05, partial η2 = .016 and there was a significant main effect for
country F (1,400) = 62.146 p < .001, partial η2= .134, which confirmed that the Indonesian sample
(M = .307) seemed to be more likely to join a protest for a noble/ social cause that UK participants
(M = -.473), see Figure 7.9 which further illustrates the results.
There was no significant main effect for gender F (1,400) = .023 p > .05, partial η2 = .00, there
was also no significant main effect age F (1,400) = 2.473 p < .05, partial η2 = .006.
-0.4
-0.2
0
0.2
0.4
UK Indonesia
Mea
n r
atin
gs
Priming
Figure 7.8. Graphical illustration of the interaction effect of country and priming (Scenerio 3)
Individualism Collectivism
182
7.3.9.5. When I am on Facebook when I see my friends talk about their achievements. I would
congratulate them. How likely are you to act in a particular way?
There was no significant main effect for priming F (1,400) = .552 p < .05, partial η2 = .001,
which failed to support the differences in responses due to cultural orientation. There was no
significant main effect for country F (1,400) = 1.410 p > .05, partial η2 = .004 which failed to
establish differences in responses due to geographical differences and there was also no significant
interaction between priming and country F (1,400) = .406, p > .05, partial η2 = .001.
There was a significant main effect for gender F (1,400) = 8.353 p < .05, partial η2 = .020.
however, there was no significant main effect for cultural orientation F (1,400) = .769 p > .05,
partial η2 = .002, for age F (1,400) = 1.509 p > .05, partial η2 = .004.
7.3.9.6. While on Facebook I often seem to join social groups that has personal importance
to me. How likely are you to act in a particular way?
There was no significant main effect for priming F (1,400) = .016 p > .05, partial η2 = .00 which
failed to support the influence of culture on participant responses which is also reflected in the
effect size. There was also no significant interaction between priming and country F (1,400) =
-0.6
-0.4
-0.2
0
0.2
0.4
UK Indonesia
Mea
n r
atin
gs
Priming
Figure 7.9. Graphical illustration of the interaction effect of country and priming (Scenerio 4)
Individualism Collectivism
183
.006, p > .05, partial η2 = .00. There was a significant main effect for difference in country F
(1,400) = 23.648 p < .001, partial η2 = .056, which suggest that difference in country did have an
impact on the ratings of the participants. Indonesian sample (M = .20) seemed to be more likely to
join social groups that has personal importance to them than the UK participants (M = -.31). This
is also illustrated in Figure 7.10. as the responses are higher for the Indonesian sample in both the
priming conditions and again there seems to be an increase in scores when the UK participants are
primed with collectivist self-construal suggesting that members with collectivist self-construal are
more likely to become members of other social groups. This is also similar to the findings of “iv)
While on Facebook I get a request to join a protest for a noble/ social cause. How likely would you
join the protest?” above. Both these findings are supportive of group behaviour which supports
collectivist behaviour.
There was however no significant main effect for cultural orientation F (1,400) = .110, p > .05,
partial η2 = .00, for gender F (1,400) = .447 p > .05, partial η2 = .001, for age F (1,400) = .129 p <
.05, partial η2 = .00,
-0.4
-0.3
-0.2
-0.1
0
0.1
0.2
0.3
UK IndonesiaMea
n r
atin
gs
Priming
Figure 7.10. Graphical illustration of the interaction effect of country and priming (Scenerio 6)
Individualism Collectivism
184
7.4. Summary
This section will summarise the findings of Study 4 and then the next section will compare the
results between the studies. Table 7.3 below summarizes the hypotheses that in Study 4 and their
results.
Table 7.3.
Illustrates the results of the hypotheses in Study 4 and its outcomes
DV
Hypotheses
() / (X)
TST
H12: Participants will respond more strongly when they are primed
with their consistent cultural self then when they are primed with
inconsistent cultural self, such that members from collectivist
countries will score higher on levels of collectivism when they are
primed with collectivism self-construal than when they are primed
with individualism self-construal. Similarly, when members from
individualist countries are primed with individualism self-construal
they will score higher on the level of individualism than levels of
collectivism.
Group
Identity
H13: Priming would have a positive effect on group identification
when primed with consistent cultural orientation than when primed
X
185
with inconsistent cultural orientation. Furthermore, Indonesian
sample would score higher on group identification than UK sample.
Self-
enhancement
H14: Priming would have an effect on how members rated themselves
when compared with others on cultural attributes, such that it helps
them to self-enhance. It is further anticipated that members from
both cultures would rate themselves higher on their cultural traits.
(Indonesia)
X (UK)
Perceived
Social
Support
H15: Priming would have a significant effect on social support when
members are primed with consistent cultural orientation than when
they are primed with inconsistent cultural orientation.
X
Trust
H16: Priming would have an effect on online trust. It is hypothesized
that when members are primed with collectivism condition then it
would increase online trust as it is anticipated that Indonesian
participants would score higher in level of online trust than the UK
participants. Alternatively, when members are primed with
individualism condition then it would reduce their level of online trust
for their social contacts.
Priming
However,
failed to
support H16
Privacy
H17: Priming would have an effect on levels of online privacy concern
such that members from collectivist countries would score higher on
the scores on privacy concern than members from individualist
countries.
Priming
However,
failed to
support H17
186
Study 4 was conducted with an aim to confirm whether self-construal priming could influence
participants’ responses to the Facebook-related outcome measures. This would confirm the causal
role of culture on the online behaviours and psychological states of individuals. This section is a
brief compilation of the results in Study 4. First of all, results for the Ten-Statements-Test
confirmed that priming had an effect on participants’ self-construal. Overall participants scored
less, which meant that they promoted individualist self-descriptions when primed with
independent self-construal than when primed with interdependent self-construal. When UK
participants were primed with individualism self-construal, they scored lower than Indonesian
participants. When the UK participants were primed with collectivism self-construal they scored
lower than the Indonesian participants as expected, however their scored seemed to improve when
they were primed with collectivist self-construal.
Online Group Identity
Although priming did not have a significant effect on online group identity, country difference
confirmed that participants in the UK identified with their group members more than the
Indonesian participants.
Online self enhancement
The effect of priming was significant for collectivist attributes (see Table 7.3). For the collectivist
attributes participants in the UK scored themselves lower on collectivist attributes than Indonesian
participants. This was as expected that members from Indonesia would advocate attributes that
supports collectivist behaviour and members from UK would advocate attributes that supported
individualist behaviour. However, participants in the UK also scored themselves lower on
individualist attributes than collectivist attributes.
Perceived Online Social Support
Priming did not have any effect on the scores of perceived online social support (see Table 7.3)
which failed to support the causal role of culture on participant responses.
187
Online Trust
Priming did have an effect on how much members trust their social network on SNS (see Table
7.3). When participants were primed with individualism self-construal, they scored higher on the
levels of online trust than when participants were primed with collectivism self-construal.
Additionally, participants in the UK trust their social contacts more than the Indonesian
participants.
Online Privacy concern
Priming did have a significant effect on participant scores (see Table 7.3). Results confirm that
participants who were primed with individualism self-construal showed higher levels of privacy
concern than participants who were primed with collectivism self-construal. Furthermore, country
difference confirmed that participants in the UK showed higher levels of privacy concern than
participants in Indonesia which was in contrary to the findings in Study 2. However, this study did
confirm the effect of priming.
Scenario Task
I log on my Facebook and receive a friend request from my family member. How likely is
that you will accept the friend request?
There was no significant effect of priming as such the differences on responses among UK and
Indonesian participants could not be attributed to cultural differences.
I log on my Facebook and receive a friend request from an unknown person. How likely will
you accept the friend request?
There was a significant main effect of priming and the results confirmed that priming members
with individualism self-construal, did not affect their ratings on whether members would accept
friend request from an unknown person. However, when Indonesian participants were primed with
individualism self-construal they would be more likely to accept a friend request from an unknown
person than when they were primed with collectivism self-construal.
188
While logging on Facebook I am reminded to update my privacy settings. How likely would
you update your privacy settings?
There was a significant main effect for priming and the results confirmed that individualist self-
construal reduces the likelihood of updating privacy settings when reminded than collectivist self-
construal, which suggests that members from collectivist countries when reminded are likely to
update privacy settings to feel protected than members from individualist countries.
While on Facebook I get a request to join a protest for a noble/ social cause. How likely would
you join the protest?
There was no significant main effect for priming as such the differences in participant responses
could not be attributed to cultural differences. However, there was a significant main effect for
country and cultural orientation which confirmed that Indonesian sample were more likely to join
a protest for a noble/ social cause.
When I am on Facebook when I see my friends talk about their achievements. I would
congratulate them. How likely are you to act in a particular way?
There was no significant main effect for priming as such the differences in participant responses
could not be attributed to cultural differences.
While on Facebook I often seem to join social groups that has personal importance to me.
How likely are you to act in a particular way?
There was no significant main effect for priming however there was a significant main effect for
country which confirmed that Indonesian participants were more likely to join social groups that
had personal importance to them.
7.5. Comparison of Results
Online Group Identity
Study 2 (see section 5.4, Chapter 5) confirmed that the Indonesian sample identified with their
group more than the UK sample and that by being present on SNS members can increase their
189
group identification and not by what type of activity they did and the results were quite similar for
members in both cultures. Study 4 further confirmed the causal role of culture as it was seen that
members from individualist countries identified with their social network more than the members
from collectivist countries which was not what was found in Study 2. However, it is important to
note that the difference could be due to the measurement scale and the sample size itself.
Online self-enhancement
Study 2 (see section 5.5, Chapter 5) confirmed that use of SNS, like Facebook decreased levels
of self-enhancement in both cultural orientations. Furthermore, the results also confirmed that
increase in individualism decreased self-enhancement. As expected members from both cultural
orientations self-enhanced on cultural specific attributes but there was a decrease in level of self-
enhancement. Study 4 however failed to support the causal role of culture on the differences in
responses in both countries as it was seen that in Study 4, when participants were primed with
collectivist attributes, Indonesian participants scored higher than participants in the UK, which
was as anticipated. However, when members were primed with individualist attributes,
participants in Indonesia scored higher than participants in UK, suggesting that Indonesian
members scored themselves higher during both priming conditions.
Study 2 confirmed that members from both countries rated themselves higher on collectivist
attributes than individualist attributes. While this was representative of Indonesian sample, higher
levels of collectivism were not expected from participants in UK. The difference in scores could
be due to the measuring scales used, the sample size or could even be attributed to the level of
anonymity on SNS which makes members behave in ways that might not necessarily be consistent
with their traditional cultural norms.
Online perceived social support
While Study 2 confirmed that both type of activities and frequency of use of SNS, like Facebook
increased perceived social support in both cultures. However, the causal role of culture could not
be established through the priming manipulation. Again, this could be due to the measurement
190
scales used or the results may be affected due to the difference in sample size. Although there
exists a difference in the perception of online social support due to how members used SNS (see
sub section 5.6, Chapter 5) the results attained failed to conclude that this difference was due to
the influence of cultural orientation of its members.
Online Trust
Study 4 confirms that priming had an impact on online behaviour however priming members
with individualism increased levels of trust on SNS than priming with collectivism condition and
it was seen that trust for social group was higher among UK participants than among Indonesian
participants.
Study 2 confirmed that the type of activities (see sub section 5.6, Chapter 5) increased the level of
trust among members.
Study 1 (see sub section 4.3.3, Chapter 4) however confirmed that there was a higher level of trust
among Indonesian participants than among UK participants.
However, Study 4 confirmed that there was a higher level of trust for their online social group
among UK participants than among Indonesian participants. This confirms the differences in
cultural behaviour on how members trust others on SNS.
While use of SNS help to increase trust among members from both cultures, higher levels of
individualism have been seen to increase trust among its social contacts.
While it would be expected that as members from collectivist countries would have higher number
of social contacts with whom they would have established offline relationship as they would prefer
maintaining their social circle of contacts, there would exist a higher level of trust among people
from collectivist countries than members from individualist countries, this was contrary to what
the results. Alternatively, it could also be argued that when members become familiar about the
behaviour of others, it helps to reduce a feeling of uncertainly, however it also helps members to
understand others more due to which probably members from collecti8vist countries scored lower
on group identity, which further confirms that maintaining group membership is more important
hence remain part of certain groups but it does not necessarily mean that they trust their group
191
members. However, the differences in behaviour on how members trust others on SNS is certainly
influenced by differences in cultural orientation.
Online Privacy concern
Priming did seem to have an influence on cultural orientation on members which supports the
findings in Study 2 as the differences in behaviour between members in UK and Indonesia is due
to the causal role of culture. This was also further supported as it was seen that country difference
had an effect on the how members perceived online privacy. The results also confirm that increase
in individualism cultural orientation increased privacy concern among its members in Study 3 (see
Chapter 7). However, this was quite the opposite in Study 2 (see sub section 5.8, Chapter 5) as the
confirmed that the Indonesian sample were more concerned about their privacy online, although
the effect was small. Although the results also confirmed that members from individualist
countries showed concern about their privacy online, it was less than collectivist members. This
difference in results could be attributed to the difference in sample size or even due to the reduced
version of the online survey scale that might have failed to capture participant responses in its
entirety.
Furthermore, Study 2 also confirmed that increase in activities and consistency of being present
on SNS increased privacy concern for members with individualist cultural orientation, however
only by browsing through SNS members became more cautious about their privacy among
collectivist cultures. This certainly highlights the fact that the more members become aware of
other’s information, the more cautious they become. Alternatively, it could also be argued that
there exists a better education facility in Western countries than among Eastern countries, due to
which people might not be as concerned about the online risks.
192
Chapter Eight
Discussion
8.1. Overview of findings
This chapter begins with the aim and overview of the research, followed by a discussion of the
results in line with the hypotheses. In addition, the findings of the models, limitations and
implications and future recommendations of the study are also discussed in the later sections.
The aim of this research was to conduct a cross-cultural investigation to evaluate the effect of
culture on the decisions, behaviours and attitudes people display online. While the study of human
behaviour, and the study of the variables selected for this project, is at the heart of psychology,
this research aimed to extend such efforts to an online environment based on the unprecedented
popularity of SNS globally, with both benefits (Ellison et al., 2007; Myers, 2000) as well as adverse
implications (Kuss & Griffiths, 2011; Kuss et al., 2013) documented by researchers. Facebook was
used as the platform for understanding online communication behaviour across cultures, i.e., in the
UK (representing an individualist culture), India and in Indonesia (both representing collectivist
cultures). The research consisted of four studies. Study 1 was an online survey focused on scale
development, Study 2 was the refined version from Study 1 and allowed for advanced model
building, Study 3 was an experimental study with the aim of calibrating a priming manipulation
and finally Study 4 was a developed and refined version of Study 3 looking at the effects of cultural
priming on the core outcome variables. Using SEM models from the survey data and data from
the experimental study, the results provided an insight how culture influences decisions and actions
people take in an online environment with consequences at an interpersonal level and at a group
level through a survey and priming experiment.
Online Social Identity
Self-reported scores on online group identity, which looked at how much members related to
their social group suggests that the time members invest on Facebook, i.e., through the hours they
spend together and the number of days members logged consistently online had an effect on the
identification with their group members. SEM models confirm that for the collectivists and
193
individualists, number of days and number of hours’ members used Facebook increased group
identity and not necessarily on type of activities which partially supported “H1: Facebook activities
(type) and frequency (time) will mediate the relationship between culture and group identity such
that increase in Facebook activity will increase the level of group identity”. This is in line with SIT
(Lea & Spears, 1991), where members rely on being available on SNS, which shows that they are
loyal to their social network, irrespective whether they are actually contributing towards group
goals or not.
Even though members are identifiable through their online profiles, there still exists some level
of anonymity due to the lack of physical cues, which help members to openly relate to other group
members, thereby resulting in stereotypical behaviours as they agree to group norms (Spears et al.,
1990). Identification with the group members is crucial so that members can develop a social
identity with the group (see SIT & SIDE). Furthermore, Indonesian participants reported a higher
number of offline social contacts online, which would force them to behave in stereotypical ways
(Douglas & McGarty, 2001) helping them maintain group harmony and giving them opportunity
to self-enhance (Noel et al., 1995) as they might also be able to voice their opinions online more
easily which might not have been possible before. On this basis, it is plausible why Indonesian
participants identified with their social contacts more strongly than UK participants in Study 2
further supporting “H2: Group identification would be stronger among members from collectivist
cultures than members from individualist cultures.”. However, this was contrary to the results that
were found in Study 4, where priming was used to manipulate cultural orientation, where
participants in the UK identified with their social groups more strongly than the Indonesian
participants. This failed to support “H13: Priming would have a significant positive effect on group
identification when primed with consistent cultural orientation than when primed with inconsistent
cultural orientation. Furthermore, Indonesian sample would score higher on group identification
than UK sample” in Study 4 Although both results are contradicting to each other, it points out
that in both cultures identification is a crucial requirement to develop one’s social identity in an
online world.
The type of Facebook activities did not help to establish group identification on SNS and did
not act as a mediator between cultural orientation and identification. In the models in Study 3, the
direct path from Facebook frequency (FB Days and FB HR) were quite similar for both cultural
194
orientations suggesting that dedicating time to use Facebook would help members from both
cultures to identify with other members in the group. This could be easy for members from
Individualist countries as they have been seen to spend more time on Facebook than the Indonesian
sample which is also in line with past findings (ref). It is however argued that this difference could
be due to factors influenced by the availability and access of the internet, social responsibilities or
even social restrictions that might cause such difference in the use of Facebook.
Online Self-Enhancement
Self-enhancement scores were based on comparing the difference in ratings on cultural specific
attributes given by the participants for themselves in comparison to their Facebook social contacts.
In Study 4, the priming conditions did not have an effect on participant responses, thereby failing
to support “H14: Priming would have a significant effect on how members rated themselves when
compared with others on cultural attributes, such that it helps them to self-enhance. It is further
anticipated that members from both cultures would rate themselves higher on their cultural traits”.
However, Study 4 confirmed that members from both collectivists and individualist countries self-
enhanced more on collectivists attributes than on individualist attributes on SNS, when they are
compared with others. This was an interesting highlight in the study as although the collectivists
self-enhanced on their collectivist attributes, members from individualist culture interesting scored
higher on collectivist attributes too unlike past research (Sedikides et al., 2005; Sedikides et al.,
2007). However, while the Indonesian participants scored lower on individualist attributes than
UK participants, as expected, Indonesians also scored lower on collectivist attributes than the UK
participants which was unexpected. This had highlighted the influence of culture on how
individuals perceive not just others but themselves too. As members from collectivist cultures,
Indonesian participants avoided direct and open self-promotion and gave themselves lower ratings.
Similarly, as UK participants come from individualist culture, they exhibited more direct self-
promotion which led to stronger endorsement of attributes. However, as this was a comparison
rating, the UK participants gave themselves higher ratings in both cultural attributes. These
findings partly contradict previous claims (Heine & Lehman, 1999; Heine & Hamamura, 2007;
Kitayama et al., 1997) that East Asians only self-enhance by self-criticism, however supports the
claim that self-enhancement is an universal concept (Sedikides et al., 2003; Sedikides et al., 2005).
195
Social interactions on SNS attract members because of assumed psychological benefits that
possibly result from such interactions. However, interactions on Facebook reduced feelings of self-
enhancement in both the cultures in the models built in Study 2, which fails to support “H4:
Facebook activities and Facebook frequency will mediate the relationship between cultural
orientation of participants and their online self-enhancement such that increase in Facebook
activities and FB frequency will increase self-enhancement in both cultures”. This is in contrast to
past findings (Valkenburg et al., 2006; Zywica & Danowski, 2008). However, these past studies
had also highlighted that negative feedback could also damage self-esteem and well-being. It could
be a possibility that members are concerned about the negative feedback they might receive due
to which they don’t rely on the feeling of self-enhancement as predicted. Furthermore, even having
social contacts on SNS, like Facebook failed to result in self-enhancement which failed to support
“H5: There will be a positive relationship between number of Facebook friends’ participants have
in both cultures and online self enhancement such that increase in number of social contacts on
Facebook will increase feelings of self-enhancement for participants in both UK and Indonesia”.
Perceived Online Social Support
In Study 2, the results supported “H6: Facebook activities (time and frequency) would mediate
the relationship between culture and social support, such that an increase in user behaviour
(Facebook activities and book Facebook frequency) would increase perceived social support for
its members in both cultures.” and increase in Facebook activities increased perceived online social
support for both individualists and collectivists. However, priming failed to have any effect on the
scores of individuals which failed to support “H15: Priming would have a significant effect on
social support when members are primed with consistent cultural orientation than when they are
primed with inconsistent cultural orientation” as the results in Study 4 found that there was no
effect of priming or country on the social support scores of participants across both cultures.
Interestingly, the results failed to support “H7: Increase in the number of social contacts on
Facebook would have an impact on perceived social support that members receive online” in Study
2 as having high numbers of social contacts on SNS did not result in receiving social support as
although the Indonesian sample had higher number of social contacts than the UK sample, they
196
scored lower on levels of social support. However, the results highlight the fact that interactions
on SNS does lead to social support which has its benefits (Ellison et al., 2007; Valenzuela et al.,
2009), however the role of cultural differences cannot be ignored. It could also be possible that
members from collectivist cultures are anxious of getting negative feedback which can have
adverse impact (Valkenburg et al., 2006) hence do not seek social support from their social
contacts online. It could be argued that they fear that others might come to know that they are
seeking social support from others which might also have adverse impact on their self-esteem. It
is further argued that this is a compensatory behaviour for members who might not be able to seek
online social support as even then such online interaction helps them to meet people online with
whom they identify with (McKenna & Bargh, 2000). SNS can thus be seen to become a platform
where members are not only able to maintain old ties and make new ones (Ellison et al., 2007),but
it can also keep members happy (Myers, 2000). Such findings are especially crucial for individuals
who have difficulty in establishing offline relationships as such support can help boost their self-
esteem (Gonzales & Hancock, 2011). However, this study has highlighted that having a high
number of social contacts on SNS might not always be providing social support and might be
having a negative impact (Tong et al., 2008) as seen in the case of Indonesians as they scored
lower on perceived social support which somewhat contradicts past findings (Myers, 2000) as even
though Indonesian participants had higher number of online bonds, they scored less on social
support than the UK participants. On the contrary (Kim, Sohn, & Choi, 2011) suggest that
collectivists mainly use SNS to secure social support whereas individualists seek entertainment. It
is therefore argued that in order to maintain potential online relationships who can provide social
support in times of need, members need to have a truthful image of oneself and they should
maintain a realistic number of friends on SNS which can vary between cultures. SNS can thus help
members to display their social self rather than their personal self by displaying their profile
information which forms their social identity benefitting members who might not always have the
confidence to interact with others openly (Tajfel & Turner, 1979; Tajfel, 1978; Turner et al., 1989).
Furthermore, the editable functionality on SNS is argued can also help members to provide
information that displays group orientated behaviour (Klein et al., 2007) which can help members
to develop a trusting relationship which they find it difficult to get offline. It might be due to this
reason that both individualists and collectivists scored themselves higher on collectivist attributed
which promotes group oriented behaviour as seen in Study 4. However the maintenance of trusting
197
relationship depends on providing a truthful representation of the members that is provided online
(Gibbs et al., 2006) where members not only get subjective well-being of themselves but also helps
members to receive more social support (Ko & Kuo, 2009; Zhao et al., 2008) and as such can also
equip members to deal with negativities in life (Dunn et al., 2007; Steele, 1988). This further
highlights that although SNS is useful in providing social support to its members, a truthful
representation of oneself is requited which promotes group behaviour. It is argued that although
the Indonesian participants scored lower on perceived online social support scale, it also highlights
the importance of the online platform as a potential and successful medium to provide social
support to their members.
Online Trust
SNS, like Facebook is a platform that helps members develop trust among its members which
is crucial in an online environment (Jarvenpaa & Leidner, 1998) due to the level of anonymity that
comes with it. Such level of trust would be more among social groups that have strong in-group
bonding. This might be easier among members who might already know their social contacts
offline. The online world allows its members to be exposed to a world which equips them to reach
out to new friends and as such it is not always possible to have established offline relationship with
their social contacts. While the development of online relationship is an extension of the offline
relationship (Lampe et al., 2006), online behaviour of members would vary due to their cultural
selves. The results in Study 4 partly supported “H16: Priming would have an effect on online trust.
It is hypothesized that when members are primed with collectivism condition then it would
increase online trust as it is anticipated that Indonesian participants would score higher in level of
online trust than the UK participants. Alternatively, when members are primed with individualism
condition then it would reduce their level of online trust for their social contacts”. Interestingly,
the results confirmed that cultural orientation did have an impact on how individuals maintain
trusting relationships online as having individualist attributes had been seen to increase the level
of trust among members online. While SNS, like Facebook promotes individualist behaviour
where members have the freedom to express their selves from their perspective, it can thus be
argued that due to the level of anonymity that exists in such online environment, members promote
their individualist selves to attract other members. It certainly highlights that when members are
198
able to portray their individualist selves, it leads to higher online trust than when they would
portray their collectivist selves. It is however argued that there needs to be a balance of how
members use their cultural selves as Study 1 (see sub section 4.33) confirmed that online trust was
higher among members from collectivist countries than individualist countries.
Development of trust also helps when members show their involvement in group tasks. This
was also partially supported in “H8. Activities and the frequency of Facebook use will mediate the
relationship between culture and online trust, such that higher use of Facebook (activities and
frequency) will increase online trust in both cultures” as the results confirmed that the activities
that members undertook increased trust in both cultures and not how much time they spent on
SNS. By being involved in the online activities members can portray their involvement and with
time that helps in the development of trust among each other (Wilson et al., 2006) further helping
them to establish their online communication. Furthermore, by being able to highlight their risk
taking attitude online (Fogel & Nehmad, 2009), members are able to reduce perception of potential
risks from other members as they are able to collect information about them (Metzger, 2004),
thereby helping to develop a mutual trusting bond.
While this study confirmed that members from collectivist countries received less social
support online than members from individualist countries could also be a reason why members
from collectivist countries had less trust for their social contacts which failed to support “H9:
Levels of trust would differ due to cultural differences such that collectivists would score higher
on the level of trust for their members than individualists”. This falls in line with the findings by
(Kim et al., 2011) and therefore, such findings will have real life implications as social support on
SNS will help members enhance trusting relationships among social groups in an online
environment which might be really crucial for members from collectivist countries. The activities
members indulge in helps to persuade their group members that they are loyal, which helps to
strengthen their identity performance (Klein, Spears, & Reicher, 2007) further helping them to
accentuate their group membership (Lea et al., 2001; Postmes et al., 2001; Sassenberg & Postmes,
2002). Due to the level of anonymity online, such trusting interpersonal and intrapersonal
relationships are crucial as there exists an agreement of reciprocity among the members. Such
reciprocity (Yamagishi & Kiyonari, 2000) is therefore argued would be stronger when there exists
a strong trusting bond, which based on the findings in this study is argued to be stronger among
199
members from individualist cultures than members from collectivist cultures, which might explain
the reason for more passive interaction on SNS among collectivist social groups. Interestingly, as
discussed before when the positive group identity is highlighted, members act in accordance to
their cultural group norms. This was supported in the findings from the ratings on the “scenarios”.
For example, when asked “While on Facebook I get a request to join a protest for a noble/ social
cause. How likely would you join the protest?”, members from collectivist countries were more
likely to join protest for a noble or a social cause. This further highlights the importance of cultural
values on SNS as it promotes group behaviour and social responsibility (Markus & Kitayama,
1991). Additionally, when asked “While on Facebook I often seem to join social groups that has
personal importance to me. How likely are you to act in a particular way?” members from
collectivist countries were more likely to join social groups that had personal importance to them.
This however not only confirms the interdependent self-construal of members from collectivist
countries but also highlights that there exists a feeling of independence too on SNS among such
members. It is argued that such result could be due to the fact that there exists some level of
identifiability among members from on SNS which makes them feel more accountable (Douglas
& McGarty, 2001). Furthermore, such behaviour also helps members to display their personal
profile among others which helps them to voice their opinions which they not always be able to
offline.
Online Privacy Concern
The results highlight that privacy concern was evident among members from both cultures.
While Study 2 confirmed that members from collectivist cultures were more concerned about their
privacy concern and supported “H11: Privacy concern will be more explicit among members from
collectivist cultures than members from individualist cultures, such that collectivists would score
higher in their privacy concern than individualists”. On the contrary Study 4 confirmed that privacy
concern was higher among members from individualist countries than members from collectivist
countries which partially failed to support”H17: Priming would have an effect on levels of online
privacy concern such that members from collectivist countries would score higher on the scores
on privacy concern than members from individualist countries” and cultural orientation did have
an effect as individualist attributes increased online privacy concern than collectivist attributes.
200
This could be due to the fact that there exist better education facilities available from various bodies
that help members of public to be educated on the risks of online communication unlike countries
in collectivist cultures where such facilities are still developing. However, when asked “7.3.10.3.
While logging on Facebook I am reminded to update my privacy settings. How likely would you
update your privacy settings?”, the results confirmed that collectivism increases the likelihood of
updating privacy settings. Additionally, when members were asked “7.3.10.2. I log on my
Facebook and receive a friend request from an unknown person. How likely will you accept the
friend request?”, the results confirm that UK participants were more inclined to accept friend
request from unknown person however increase in collectivism reduces this tendency. These
findings highlight that in general collectivist attributes makes members become cautious about
their online information and privacy risks which is in line with their cultural behaviour. While
country differences highlight that there exist differences in user behaviour and tendency of
members to ignore potential online risks associated with an interaction (Gross & Acquisti, 2005)
due to the benefits that members achieve with such communication (Bender et al., 2011; Boyd,
2007; Culnan & Armstrong, 1999; Dinev & Hart, 2006; Lampe et al., 2007). While the results also
highlight that participants in the UK were less likely to update privacy settings when reminded,
also heightens the social responsibility for SNS providers to enhance privacy options for its
members. Additionally, this study also calls for collectivists countries like Indonesia to enhance
their education among people regarding online risks.
In order to have successful communication on SNS members need to involve in various
activities to establish a connectivity with others in the group. The results confirmed that activities
and frequency of use of SNS would increase privacy concern among members which partially
supported “H10: There would be a negative relationship between Facebook activities (active &
passive), Facebook frequency (days and hours spent) and online privacy concern, such that
increase in Facebook activities and Facebook frequency would decrease privacy concern among
online users”. It is argued that as members spend time online they collect information about other
members which makes them predict their behaviours, thereby increasing their privacy concern. It
is important to highlight the fact that the SNS equips members to be able to judge and decide the
pros and cons of an interaction (Homans, 1958). While SNS provides the opportunity to its
members to know about the personal lives of others which might not has always been possible
before it is argued that such surveillance would only be possible when there are less restrictions as
201
members would be able to freely view information about others. It might be due to this reason that
although members from individualist cultures are aware of the privacy risks they are less likely to
update their privacy settings as it allows others to view their information and also allows them to
view others information without restrictions (Gross & Acquisti, 2005) and build trust among each
other (Culnan & Armstrong, 1999) as while the motive among individualist members is to entertain
themselves whereas the motive among collectivists is to maintain their social ties through online
communication (Kim, Sohn, & Choi, 2011). Being able to collect information about other
members helps in identification with them which helps in depersonalization (Turner et al., 1987).
However, while it is argued that depersonalization would lead to group bias, this study showed
that group identification was more among the members from collectivist culture which highlights
that group bias could be experienced higher among collectivist social groups than individualist
social groups on SNS.
The high number of social contacts on Facebook among the Indonesian sample, suggest that
while the social network among the Indonesian sample was more diverse than the UK sample
which can potentially lead to online social tension especially if the social network has higher
number of close family contacts (Binder et al., 2009) social identification and use of filter settings
would possibly help to prevent such online tension which could be attained by increase in
education about online risks and prevention in such countries. This would be crucial in a
collectivist group as maintaining group harmony would be of prime importance to their self-
identity (Markus & Kitayama, 1991; Singelis, 1994).
8.2. Limitations
There had been unexpected challenges during data collection in India which had delayed the
overall research process (see Chapter Three) that has prevented the inclusion of longitudinal data
analysis. Therefore, future cross-cultural research should take such delays into account as
unexpected delays in data collection could have an adverse impact on the researcher and can also
have adverse impact on the research project. The cultural self-construal scale had a medium scale
reliability which might have had an impact on the results. It is therefore advised that when using
validated scales from past research to check the relevance of each item in the particular cultures
202
under investigation. This could be achieved by running a short focus group or a pilot study where
participants can discuss whether they could relate to the items in the scale.
While this research did attempt to understand the effects of both cultural orientations
(individualism and collectivism) at the same time, the results suggested that overall models
containing only one dimension were better in capturing significant responses. However, some of
the combined models also showed significant mediating effects of Facebook activities and
Facebook usage. It should be noted that effects were mostly significant for individualist self-
construal. This could be due to the fact although the survey items used in the cultural self-construal
scale (Singelis, 1994) were from past validated study, their relevance at the present time has
probably declined, which calls for such a scale to be improved based on current social situation. It
might be that due to this reason, participants were not able to entirely relate to the situations that
were given on the scale and this might have had an impact on the ratings. Furthermore, this was
also supported by moderate scale reliability of the scale items too.
Furthermore, while this was a cross-cultural study understanding the impact of each
psychological variable was important to fully understand differences in behaviour in both cultures.
While attempt was made to analyse the data collected using four different mediators, it is important
to highlight that all four mediators together make one construct, i.e., Facebook engagement. It was
difficult to single out effects of the mediators in the models possibly because there exists a
correlation between the mediators and as such future research should take its effect into
consideration while running similar analysis.
8.3 Contributions of the Study
The current research highlights some important contributes to the existing body of literature on
culture and communication on social media. This study evaluated the role of culture cross-
culturally on several psychological variables at the same time, which might have not been
attempted before. Besides this study extended the concept of SIT, SCT and SIDE to explore the
cultural dimensions in-depth. The study explored the use of several mediators to explore the effect
too. Furthermore, while there has been a lot of research focusing on data collected from Western
203
population, this study was able to address this issue by using data collected from native participants
from individualist and collectivist countries.
While the use of SNS is a global phenomenon and has become part of our lives, its use is not
restricted to any particular culture. This cross-cultural research has been able to shed light on cross-
cultural usage and activities in different cultures. While this study has successfully highlighted
how use of social media varies across cultures and affects behaviour of its members, it also
highlights the importance of understanding these cross-cultural changes over time as new
generations develop over time who are digital natives, who are now growing up with Facebook
which is experiencing rapid changes in privacy and usage provisions.
8.4. Implications and Future Research
The findings of this research sheds lights on the cultural influence on the attitudes and
behaviours of individuals in an online environment. While the above findings have been successful
in highlighting the role of culture in line with the SIT, SCT and SIDE, the present study failed to
attain models with included all the four mediators (FB Active, FB Passive, FB Days and FB HR)
in one single model along with both cultural orientation (collectivism and individualism) which
calls for further research.
While this research focused on evaluating the role of culture on various psychological variables,
it was however not possible to check other factors that could have also had an effect. The
globalisation of the use of SNSs, where members from both cultural dimension actively
participates in protests, demonstration and igniting mass hatred calls for further investigation. In
order to fully understand online behaviour it is therefore important to evaluate other factors that
affects members online, e.g., effect of religious beliefs on the attitudes and behaviour of members
on SNS. Additionally, while this research considered the two cultural dimensions (collectivism
and individualism) formulated by Hofstede (1980), further research should consider other cultural
dimensions in understanding cultural differences.
As Facebook, has been developed with an individualist culture in mind, the above findings can
also be useful for web developers who could consider developing an Eastern style Facebook which
204
could promote behaviours that are specific to Eastern culture. The results achieved could also help
future web developers to develop applications that might be able to cater to specific cultural group
which could assist members with privacy settings and provide cultural specific information. The
results highlight the differences in the perception of online risks across cultures and further
highlights the importance of online education among social network sites users.
The results are not only important for social researchers but would be beneficial for health
professionals and also the general population, as it not only help to understand the behaviour of
people in different cultures, it could also be beneficial for understanding various online behaviour
like online protests, internet addiction and online support. Although this research was not fully
able to understand the effect of culture on social support members receive online, it however, paves
a path for future researchers to evaluate this further as the results could be useful for development
and implementation of online focus groups and online support that could be benefit a wider
population. Additionally, the results can help web developers and marketers understand how
culture influences behaviour online, thereby can help them target their products to specific cultural
group of consumers.
8.5. Conclusion
The main objective of this cross-cultural research was to evaluate the role of culture on user
behaviour on SNS. Facebook was used as platform to collate online data due to its increasing
popularity. Self-reported data was collected from UK (representing individualist culture), India
and Indonesia (representing collectivist culture) in the form of survey and experiment. While Study
1 (scale development) and Study 2 consisted of online study, Study 3 (scale development) and
Study 4 consisted of priming experiments. The results confirm that difference in responses on
various psychological variables like online group identity, online self-enhancement, perceived
online social support, online trust and online privacy concern were evident due to the influence of
cultural orientation and country differences. It is therefore hoped that based on the findings in this
study, future researchers consider not just country differences but also differences in cultural
orientation of members online when they are evaluating online user behaviour.
205
References
Acquisti, A., & Gross, R. (2006). Imagined communities: Awareness, information sharing, and
privacy on the facebook. Privacy Enhancing Technologies, 36-58.
Akbulut, Y., & Günüç, S. (2012). Perceived social support and Facebook use among adolescents.
International Journal of Cyber Behavior, Psychology and Learning (IJCBPL), 2(1), 30-41.
Amichai-Hamburger, Y. 2005 ‘Internet minimal group paradigm’. CyberPsycychology and
Behavior, 8(2), 140-142.
Arbuckle, J. L. (2006). Amos (version 7.0)[computer program]. Chicago: SpSS,
Back, M. D., Stopfer, J. M., Vazire, S., Gaddis, S., Schmukle, S. C., Egloff, B., & Gosling, S. D.
(2010). Facebook profiles reflect actual personality, not self-idealization. Psychological
Science, 21(3), 372-374. doi:10.1177/0956797609360756
Barker, V. (2009). Older adolescents' motivations for social network site use: The influence of
gender, group identity, and collective self-esteem. CyberPsychology & Behavior, 12(2),
209-213.
Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator variable distinction in social
psychological research: Conceptual, strategic, and statistical considerations. Journal of
Personality and Social Psychology, 51(6), 1173.
Basabe, N., & Ros, M. (2005). Cultural dimensions and social behavior correlates:
Individualism-collectivism and power distance. Revue Internationale De Psychologie
Sociale, 18(1), 189-225.
206
Baumeister, F. R., & Leary, M. (1995). The Need to Belong: Desire for Interpersonal
Attachments as a Fundamental Human Motivation. Psychological Bulletin, 117(3), 497-529.
Baumeister, R. F. (1982). A self-presentational view of social phenomena. Psychological
Bulletin, 91(1), 3.
Behling, O., & Law, K. S. (2000). Translating questionnaires and other research instruments:
Problems and solutions Sage.
Bender, J. L., Jimenez-Marroquin, M., & Jadad, A. R. (2011a). Seeking support on Facebook: A
content analysis of breast cancer groups. Journal of Medical Internet Research, 13(1), e16.
Bentler, P. M. (1990). Comparative fit indexes in structural models. Psychological Bulletin,
107(2), 238.
Berry, J. W. (1989). Imposed etics-emics-derived etics: The operationalization of a compelling
idea. International Journal of Psychology, 24(6), 721-735.
Binder, J., Howes, A., & Sutcliffe, A. (2009). The problem of conflicting social spheres: Effects
of network structure on experienced tension in social network sites. Proceedings of the
SIGCHI Conference on Human Factors in Computing Systems, 965-974.
Blacker, D., & Endicott, J. (2002). Psychometric properties: Concepts of reliability and validity.
Handbook of Psychiatric Measures.Washington: American Psychiatric Association, , 7-14.
207
Bochner, S. (1994). Cross-cultural differences in the self-concept a test of Hofstede's
individualism/collectivism distinction. Journal of Cross-Cultural Psychology, 25(2), 273-
283.
Boyd, D. (2007). Why youth (heart) social network sites: The role of networked publics in
teenage social life. MacArthur Foundation Series on Digital learning–Youth, Identity, and
Digital Media Volume, 119-142.
Brace, N., Kemp, R., Snelgar R. (2009). An introduction to multiple regression. In Brace, N.,
Kemp, R., Snelgar R. (Ed.), SPSS for psychologists. (4th edition ed., pp. 265) Palgrave
Macmillan.
brandwatch. (2016). Marketing: 96 amazing social media statistics and facts for 2016. Retrieved
October 10, 2016 from https://www.brandwatch.com/2016/03/96-amazing-social-media-
statistics-and-facts-for-2016/
Brewer, M. B., & Gardner, W. (1996). Who is this" we"? levels of collective identity and self-
representations. Journal of Personality and Social Psychology, 71(1), 83.
Brislin, R. W. (1970). Back-translation for cross-cultural research. Journal of Cross-Cultural
Psychology, 1(3), 185-216.
Brown, J. D. (1986). Evaluations of self and others: Self-enhancement biases in social
judgments. Social Cognition, 4(4), 353-376.
Brown, J. D., & Kobayashi, C. (2002). Self‐enhancement in Japan and America. Asian Journal
of Social Psychology, 5(3), 145-168.
208
Brown, J. D., & Mankowski, T. A. (1993). Self-esteem, mood, and self-evaluation: Changes in
mood and the way you see you. Journal of Personality and Social Psychology, 64(3), 421.
Brown, R., Condor, S., Mathews, A., Wade, G., & Williams, J. (1986). Explaining intergroup
differentiation in an industrial organization. Journal of Occupational Psychology, 59(4),
273-286.
Burke, M., Marlow, C., & Lento, T. (2010). Social network activity and social well-being.
Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, 1909-
1912.
Byrne, B. M. (2001). Structural equation modeling with AMOS, EQS, and LISREL:
Comparative approaches to testing for the factorial validity of a measuring instrument.
International Journal of Testing, 1(1), 55-86.
Cheung, C. M., & Lee, M. K. (2010). A theoretical model of intentional social action in online
social networks. Decision support systems, 49(1), 24-30.
Cross, S. E. (1995). Self-construals, coping, and stress in cross-cultural adaptation. Journal of
Cross-Cultural Psychology, 26(6), 673-697.
Culnan, M. J., & Armstrong, P. K. (1999). Information privacy concerns, procedural fairness,
and impersonal trust: An empirical investigation. Organization Science, 10(1), 104-115.
Cunningham, M. R., Roberts, A. R., Barbee, A. P., Druen, P. B., & Wu, C. (1995). " Their ideas
of beauty are, on the whole, the same as ours": Consistency and variability in the cross-
209
cultural perception of female physical attractiveness. Journal of Personality and Social
Psychology, 68(2), 261.
Dake, K. (1991). Orienting dispositions in the perception of risk an analysis of contemporary
worldviews and cultural biases. Journal of Cross-Cultural Psychology, 22(1), 61-82.
Debatin, B., Lovejoy, J. P., Horn, A., & Hughes, B. N. (2009). Facebook and online privacy:
Attitudes, behaviors, and unintended consequences. Journal of Computer‐Mediated
Communication, 15(1), 83-108.
Dinev, T., & Hart, P. (2006). An extended privacy calculus model for e-commerce transactions.
Information Systems Research, 17(1), 61-80.
Douglas, K. M., & McGarty, C. (2001). Identifiability and self‐presentation: Computer‐mediated
communication and intergroup interaction. British Journal of Social Psychology, 40(3), 399-
416.
Dunn, E. W., Biesanz, J. C., Human, L. J., & Finn, S. (2007). Misunderstanding the affective
consequences of everyday social interactions: The hidden benefits of putting one's best face
forward. Journal of Personality and Social Psychology, 92(6), 990.
Dunning, D. (1995). Trait importance and modifiability as factors influencing self-assessment
and self-enhancement motives. Personality and Social Psychology Bulletin, 21(12), 1297-
1306.
Dwyer, C., Hiltz, S., & Passerini, K. (2007). Trust and privacy concern within social networking
sites: A comparison of Facebook and MySpace. AMCIS 2007 Proceedings, , 339.
210
Ellison, N. B. (2007). Social network sites: Definition, history, and scholarship. Journal of
Computer‐Mediated Communication, 13(1), 210-230.
Ellison, N. B., Steinfield, C., & Lampe, C. (2007). The benefits of Facebook “friends:” Social
capital and college students’ use of online social network sites. Journal of Computer‐
Mediated Communication, 12(4), 1143-1168.
Ellison, N., Heino, R., & Gibbs, J. (2006). Managing impressions online: Self‐presentation
processes in the online dating environment. Journal of Computer‐Mediated Communication,
11(2), 415-441.
Ellison, N., Steinfield, C., & Lampe, C. (2006). Spatially bounded online social networks and
social capital. International Communication Association, 36(1-37)
Emanuel, L., Neil, G. J., Bevan, C., Fraser, D. S., Stevenage, S. V., Whitty, M. T., & Jamison-
Powell, S. (2014). Who am I? representing the self offline and in different online contexts.
Computers in Human Behavior, 41, 146-152.
Enrique, D. A. (2016). Internet world stats: Usage and population statistics. Retrieved September
20, 2016 from http://www.internetworldstats.com/stats3.htm
Fogel, J., & Nehmad, E. (2009). Internet social network communities: Risk taking, trust, and
privacy concerns. Computers in Human Behavior, 25(1), 153-160.
Gaertner, L., Sedikides, C., & Chang, K. (2008). On pancultural self-enhancement: Well-
adjusted Taiwanese self-enhance on personally valued traits. Journal of Cross-Cultural
Psychology, 39(4), 463.477.
211
Gardner, W. L., Gabriel, S., & Lee, A. Y. (1999). “I” value freedom, but “we” value
relationships: Self-construal priming mirrors cultural differences in judgment. Psychological
Science, 10(4), 321-326.
Gibbs, J. L., Ellison, N. B., & Heino, R. D. (2006). Self-presentation in online personals the role
of anticipated future interaction, self-disclosure, and perceived success in internet dating.
Communication Research, 33(2), 152-177.
Gonzales, A. L., & Hancock, J. T. (2011). Mirror, mirror on my facebook wall: Effects of
exposure to facebook on self-esteem. Cyberpsychology, Behavior, and Social Networking,
14(1-2), 79-83.
Granovetter, M. S. (1973). The strength of weak ties. American Journal of Sociology, , 1360-
1380.
Gross, R., & Acquisti, A. (2005). Information revelation and privacy in online social networks.
Proceedings of the 2005 ACM Workshop on Privacy in the Electronic Society, 71-80.
Gudykunst, W. B., Matsumoto, Y., Ting‐Toomey, S., Nishida, T., Kim, K., & Heyman, S.
(1996). The influence of cultural individualism‐collectivism, self construals, and individual
values on communication styles across cultures. Human Communication Research, 22(4),
510-543.
Hair, J. F., Black, W. C., Babin, B. J., Anderson, R. E., & Tatham, R. L. (2006). Multivariate
data analysis (vol. 6).
212
Halder, M., Binder, J., Stiller, J., & Gregson, M. (2016). An overview of the Challenges faced
during Cross-Cultural Research. Enquire, (8), 1-18.
Hayes, A. F. (2009). Beyond baron and kenny: Statistical mediation analysis in the new
millennium. Communication Monographs, 76(4), 408-420.
Hayes, A. F. (2012). PROCESS: A Versatile Computational Tool for Observed Variable
Mediation, Moderation, and Conditional Process Modeling,
Heine, S. J., & Dehman, D. (1997). Culture, dissonance, and self-affirmation. Personality and
Social Psychology Bulletin, 23, 389-400.
Heine, S. J., & Lehman, D. R. (1999). Culture, self-discrepancies, and self-satisfaction.
Personality and Social Psychology Bulletin, 25(8), 915-925.
Heine, S. J., Lehman, D. R., Markus, H. R., & Kitayama, S. (1999). Is there a universal need for
positive self-regard? Psychological Review, 106(4), 766.
Heine, S. J., & Hamamura, T. (2007). In search of east asian self-enhancement. Personality and
Social Psychology Review : An Official Journal of the Society for Personality and Social
Psychology, Inc, 11(1), 4-27. doi:10.1177/1088868306294587
Hofstede, G. (1980). Culture’s consequences: International differences in work-related values.
Beverley Hills, CA: Sage.
Hofstede, G. H., & Hofstede, G. (2001). Culture's consequences: Comparing values, behaviors,
institutions and organizations across nations Sage.
213
Hofstede, G., Hofstede, G. J., & Minkov, M. (1991). Cultures and organizations: Software of the
mind Citeseer.
Hofstede, G., Hofstede, G. J., & Minkov, M. (1997). Cultures and organizations McGraw Hill
New York, NY.
Hogg, M. A. (2001). A social identity theory of leadership. Personality and Social Psychology
Review, 5(3), 184-200.
Hogg, M. A., & Hardie, E. A. (1991). Social attraction, personal attraction, and self-
categorization-, A field study. Personality and Social Psychology Bulletin, 17(2), 175-180.
Hogg, M. A., & Turner, J. C. (1987). Intergroup behaviour, self‐stereotyping and the salience of
social categories. British Journal of Social Psychology, 26(4), 325-340.
Homans, G. C. (1958). Social behavior as exchange. American Journal of Sociology, , 597-606.
Hooper, D., Coughlan, J., & Mullen, M. (2008). Structural equation modelling: Guidelines for
determining model fit. Articles, , 2.
Hui, C. H., & Triandis, H. C. (1986). Individualism-collectivism a study of cross-cultural
researchers. Journal of Cross-Cultural Psychology, 17(2), 225-248.
Hui, K., Tan, B. C., & Goh, C. (2006). Online information disclosure: Motivators and
measurements. ACM Transactions on Internet Technology (TOIT), 6(4), 415-441.
Jarvenpaa, S. L., & Leidner, D. E. (1998). Communication and trust in global virtual teams.
Journal of Computer‐Mediated Communication, 3(4), 0-0.
214
Ji, Y. G., Hwangbo, H., Yi, J. S., Rau, P. P., Fang, X., & Ling, C. (2010). The influence of
cultural differences on the use of social network services and the formation of social capital.
Intl.Journal of Human–Computer Interaction, 26(11-12), 1100-1121.
Johnson, B. K., & Knobloch-Westerwick, S. (2014). Glancing up or down: Mood management
and selective social comparisons on social networking sites. Computers in Human Behavior,
41, 33-39.
Joinson, A. N. (2008). Looking at, looking up or keeping up with people?: Motives and use of
facebook. Proceedings of the SIGCHI Conference on Human Factors in Computing
Systems, 1027-1036.
Jöreskog, K. G., & Sörbom, D. (1982). Recent developments in structural equation modeling.
Journal of Marketing Research, , 404-416.
Junco, R. (2012). Too much face and not enough books: The relationship between multiple
indices of facebook use and academic performance. Computers in Human Behavior, 28(1),
187-198.
Karl, K., Peluchette, J., & Schlaegel, C. (2010). Who's posting Facebook faux pas? A cross‐
cultural examination of personality differences. International Journal of Selection and
Assessment, 18(2), 174-186.
Karl, K., Peluchette, J., & Schlagel, C. (2010). A cross-cultural examination of student attitudes
and gender differences in Facebook profile content.
215
Kiesler, S., Siegel, J., & McGuire, T. W. (1984). Social psychological aspects of computer-
mediated communication. American Psychologist, 39(10), 1123.
Kim, Y., Sohn, D., & Choi, S. M. (2011). Cultural difference in motivations for using social
network sites: A comparative study of American and Korean college students. Computers in
Human Behavior, 27(1), 365-372.
Kitayama, S., Markus, H. R., & Kurokawa, M. (2000). Culture, emotion, and well-being: Good
feelings in japan and the united states. Cognition & Emotion, 14(1), 93-124.
Kitayama, S., Markus, H. R., Matsumoto, H., & Norasakkunkit, V. (1997). Individual and
collective processes in the construction of the self: Self-enhancement in the united states and
self-criticism in japan. Journal of Personality and Social Psychology, 72(6), 1245.
Klein, O., Spears, R., & Reicher, S. (2007). Social identity performance: Extending the strategic
side of SIDE. Personality and Social Psychology Review : An Official Journal of the Society
for Personality and Social Psychology, Inc, 11(1), 28-45. doi:10.1177/1088868306294588
[doi]
Ko, H., & Kuo, F. (2009). Can blogging enhance subjective well-being through self-disclosure?
CyberPsychology & Behavior, 12(1), 75-79.
Krämer, N. C., & Winter, S. (2008). Impression management 2.0: The relationship of self-
esteem, extraversion, self-efficacy, and self-presentation within social networking sites.
Journal of Media Psychology, 20(3), 106-116.
216
Krasnova, H., Spiekermann, S., Koroleva, K., & Hildebrand, T. (2010). Online social networks:
Why we disclose. Journal of Information Technology, 25(2), 109-125.
Kuhn, M. H., & McPartland, T. S. (1954). An empirical investigation of self-attitudes. American
Sociological Review, 19(1), 68-76.
Kühnen, U., & Oyserman, D. (2002). Thinking about the self influences thinking in general:
Cognitive consequences of salient self-concept. Journal of Experimental Social Psychology,
38(5), 492-499.
Kurman, J. (2001). Self-enhancement: Is it restricted to individualistic cultures? Personality and
Social Psychology Bulletin, 27(12), 1705-1716.
Kurman, J. (2003). Why is self-enhancement low in certain collectivist cultures? an investigation
of two competing explanations. Journal of Cross-Cultural Psychology, 34(5), 496-510.
Kuss, D. J., & Griffiths, M. D. (2011). Online social networking and addiction—a review of the
psychological literature. International Journal of Environmental Research and Public
Health, 8(9), 3528-3552.
Kuss, D. J., Griffiths, M. D., & Binder, J. F. (2013). Internet addiction in students: Prevalence
and risk factors. Computers in Human Behavior, 29(3), 959-966.
Lampe, C. A., Ellison, N., & Steinfield, C. (2007). A familiar face (book): Profile elements as
signals in an online social network. Proceedings of the SIGCHI Conference on Human
Factors in Computing Systems, 435-444.
217
Lampe, C., Ellison, N. B., & Steinfield, C. (2008). Changes in use and perception of Facebook.
Proceedings of the 2008 ACM Conference on Computer Supported Cooperative Work, 721-
730.
Lampe, C., Ellison, N., & Steinfield, C. (2006). A face (book) in the crowd: Social searching vs.
social browsing. Proceedings of the 2006 20th Anniversary Conference on Computer
Supported Cooperative Work, 167-170.
Lampinen, A., Tamminen, S., & Oulasvirta, A. (2009). All my people right here, right now:
Management of group co-presence on a social networking site. Proceedings of the ACM
2009 International Conference on Supporting Group Work, 281-290.
Lea, M., & Spears, R. (1991). Computer-mediated communication, de-individuation and group
decision-making. International Journal of Man-Machine Studies, 34(2), 283-301.
Lea, M., Spears, R., & de Groot, D. (2001). Knowing me, knowing you: Anonymity effects on
social identity processes within groups. Personality and Social Psychology Bulletin, 27(5),
526-537.
Leary, M. R., & Kowalski, R. M. (1990). Impression management: A literature review and two-
component model. Psychological Bulletin, 107(1), 34.
MacCallum, R. C., Browne, M. W., & Sugawara, H. M. (1996). Power analysis and
determination of sample size for covariance structure modeling. Psychological Methods,
1(2), 130.
218
Mackie, D. M. (1986). Social identification effects in group polarization. Journal of Personality
and Social Psychology, 50(4), 720.
Marin, G., & Marin, B. V. (1991). Research with hispanic populations. Sage Publications, Inc.
Markus, H. R., & Kitayama, S. (1991). Culture and the self: Implications for cognition, emotion,
and motivation. Psychological Review, 98(2), 224.
Markus, H., Fiske, A., Nisbett, R., & Kitayama, S. (1998). The cultural matrix of social
psychology. The Handbook of Social Psychology, 2
Markus, H. R., & Kitayama, S. (2010). Cultures and selves: A cycle of mutual constitution.
Perspectives on Psychological Science : A Journal of the Association for Psychological
Science, 5(4), 420-430. doi:10.1177/1745691610375557
McGorry, S. Y. (2000). Measurement in a cross-cultural environment: Survey translation issues.
Qualitative Market Research: An International Journal, 3(2), 74-81.
McKenna, K. Y., & Bargh, J. A. (2000). Plan 9 from cyberspace: The implications of the internet
for personality and social psychology. Personality and Social Psychology Review, 4(1), 57-
75.
McNamara, N., & Parsons, H. (2016). 'Everyone here wants everyone else to get better': The role
of social identity in eating disorder recovery. British Journal of Social Psychology, 55(4),
662-680. doi: 10.1111/bjso.12161
Merton, R. K. (1968). The matthew effect in science. Science, 159(3810), 56-63.
219
Metzger, M. J. (2004). Privacy, trust, and disclosure: Exploring barriers to electronic commerce.
Journal of Computer‐Mediated Communication, 9(4), 00-00.
Meyers, L. S., Gamst, G., & Guarino, A. J. (2006). Applied multivariate research: Design and
interpretation Sage.
Myers, D. G. (2000). The funds, friends, and faith of happy people.55(1), 56-67.
Nisbett, R. E., Peng, K., Choi, I., & Norenzayan, A. (2001). Culture and systems of thought:
Holistic versus analytic cognition. Psychological Review, 108(2), 291.
Noel, J. G., Wann, D. L., & Branscombe, N. R. (1995). Peripheral ingroup membership status
and public negativity toward outgroups. Journal of Personality and Social Psychology,
68(1), 127.
Oyserman, D., & Lee, S. W. (2008). Does culture influence what and how we think? effects of
priming individualism and collectivism. Psychological Bulletin, 134(2), 311.
Parks, M. R., & Floyd, K. (1996). Making friends in cyberspace. Journal of Computer‐Mediated
Communication, 1(4), 0-0.
Park, Namsu, Kee, F. Kerk and Sebastián Valenzuela. (2009). "Being immersed in social
networking environment: Facebook groups, uses and gratifications, and social outcomes."
CyberPsychology & Behavior, 12(6), 729-733.
Pedersen, P. B. (1991). Multiculturalism as a generic approach to counseling. Journal of
Counseling & Development, 70(1), 6-12.
220
Pempek, T. A., Yermolayeva, Y. A., & Calvert, S. L. (2009). College students' social networking
experiences on facebook. Journal of Applied Developmental Psychology, 30(3), 227-238.
Ponterotto, J. G., & Casas, J. (1993). Handbook of ethnic/racial minority counseling research.
Springfield, IL: Charles C.Thomas,
Postmes, T., Spears, R., Sakhel, K., & De Groot, D. (2001). Social influence in computer-
mediated communication: The effects of anonymity on group behavior. Personality and
Social Psychology Bulletin, 27(10), 1243-1254.
Raacke, J., & Bonds-Raacke, J. (2008). MySpace and facebook: Applying the uses and
gratifications theory to exploring friend-networking sites. Cyberpsychology & Behavior,
11(2), 169-174.
Reicher, S. D., Spears, R., & Postmes, T. (1995). A social identity model of deindividuation
phenomena. European Review of Social Psychology, 6(1), 161-198.
Reicher, S., & Levine, M. (1994). Deindividuation, power relations between groups and the
expression of social identity: The effects of visibility to the out‐group. British Journal of
Social Psychology, 33(2), 145-163.
Ren, Y., Kraut, R., & Kiesler, S. (2007). Aapplying Common Identity and Bond Theory to
Design of Online Communities. Organization Studies, 28, 377-408. doi:
10.1177/0170840607076007
Roloff, M. E. (1981). Interpersonal communication: The social exchange approach Sage
Publications, Inc.
221
Rosen, P., & Sherman, P. (2006). Hedonic information systems: Acceptance of social
networking websites.
Ross, M., Heine, S. J., Wilson, A. E., & Sugimori, S. (2005). Cross-cultural discrepancies in self-
appraisals. Personality & Social Psychology Bulletin, 31(9), 1175-1188. doi:31/9/1175 [pii]
Rouis, S., Limayem, M., & Salehi-Sangari, E. (2011). Impact of facebook usage on students’
academic achievement: Role of self-regulation and trust. Electronic Journal of Research in
Educational Psychology, 9(3), 961-994.
Sassenberg, K., & Postmes, T. (2002). Cognitive and strategic processes in small groups: Effects
of anonymity of the self and anonymity of the group on social influence. British Journal of
Social Psychology, 41(3), 463-480.
Schumacker, R. E., & Lomax, R. G. (2004). A beginner's guide to structural equation modeling
Psychology Press.
Sechrest, L., Fay, T. L., & Zaidi, S. H. (1972). Problems of translation in cross-cultural research.
Journal of Cross-Cultural Psychology, 3(1), 41-56.
Sedikides, C., Gaertner, L., & Toguchi, Y. (2003). Pancultural self-enhancement. Journal of
Personality and Social Psychology, 84(1), 60.
Sedikides, C., Gaertner, L., & Vevea, J. L. (2007). Inclusion of theory‐relevant moderators yield
the same conclusions as sedikides, gaertner, and vevea (2005): A meta‐analytical reply to
heine, kitayama, and hamamura (2007). Asian Journal of Social Psychology, 10(2), 59-67.
222
Sedikides, C., Skowronski, J. J., & Gaertner, L. (2004). Self-enhancement and self-protection
motivation: From the laboratory to an evolutionary context. Journal of Cultural and
Evolutionary Psychology, 2(1-2), 61-79.
Sekaran, U. (1983). Methodological and theoretical issues and advancements in cross-cultural
research. Journal of International Business Studies, 14(2), 61-73.
Shelbourne, D. C., & Stewart, L. A. (1991). The MOS social support survey. Soc. Sci. Med.,
32(6), 705-714.
Sheldon, P. (2008). The relationship between unwillingness-to-communicate and students’
facebook use. Journal of Media Psychology, 20(2), 67-75.
Shrout, P. E., & Bolger, N. (2002). Mediation in experimental and nonexperimental studies: New
procedures and recommendations. Psychological Methods, 7(4), 422.
Siegel, J., Dubrovsky, V., Kiesler, S., & McGuire, T. W. (1986). Group processes in computer-
mediated communication. Organizational Behavior and Human Decision Processes, 37(2),
157-187.
Singelis, T. M. (1994). The measurement of independent and interdependent self-construals.
Personality and Social Psychology Bulletin, 20(5), 580-591.
Spears, R., & Lea, M. (1994). Panacea or panopticon? the hidden power in computer-mediated
communication. Communication Research, 21(4), 427-459.
223
Spears, R., Lea, M., Corneliussen, R. A., Postmes, T., & Ter Haar, W. (2002). Computer-
mediated communication as a channel for social resistance the strategic side of SIDE. Small
Group Research, 33(5), 555-574.
Spears, R., Lea, M., & Lee, S. (1990). De‐individuation and group polarization in computer‐
mediated communication. British Journal of Social Psychology, 29(2), 121-134.
Statista. (2016). Number of languages supported on leading web platforms as of 2012. Retrieved
September 20, 2016 from https://www.statista.com/statistics/245530/number-of-languages-
supported-by-web-platform/
Statista. (2016). Number of mobile monthly active Facebook users worldwide from 1st quarter
2009 to 2nd quarter 2016 (in millions). Retrieved October 10, 2016 from
https://www.statista.com/statistics/277958/number-of-mobile-active-facebook-users-
worldwide/
Statista. (2016). Percentage of global online population using Facebook as of may 2014, by
region. Retrieved October 10, 2016 from https://www.statista.com/statistics/241552/share-
of-global-population-using-facebook-by-region/
Statista. (2016). The statistics portal. Retrieved October 10, 2016 from
https://www.statista.com/statistics/272014/global-social-networks-ranked-by-number-of-
users/
Steele, C. M. (1988). The psychology of self-affirmation: Sustaining the integrity of the self.
Advances in Experimental Social Psychology, 21, 261-302.
224
Steiger, J. H. (1998). A note on multiple sample extensions of the RMSEA fit index.
Steinfield, C., Ellison, N. B., & Lampe, C. (2008). Social capital, self-esteem, and use of online
social network sites: A longitudinal analysis. Journal of Applied Developmental Psychology,
29(6), 434-445.
Subrahmanyam, K., & Greenfield, P. (2008). Online communication and adolescent
relationships. The Future of Children, 18(1), 119-146.
Tajfel, H. E. (1978). Differentiation between social groups: Studies in the social psychology of
intergroup relations. Academic Press.
Tajfel, H., Billig, M. G., Bundy, R. P., & Flament, C. (1971a). Social categorization and
intergroup behaviour. European Journal of Social Psychology, 1(2), 149-178.
Tajfel, H., & Turner, J. C. (1979). An integrative theory of intergroup conflict. The Social
Psychology of Intergroup Relations, 33(47), 74.
Takata, T. (2003). Self-enhancement and self-criticism in Japanese culture an experimental
analysis. Journal of Cross-Cultural Psychology, 34(5), 542-551.
Tanis, M., & Postmes, T. (2003). Social cues and impression formation in CMC. Journal of
Communication, 53(4), 676-693.
Tong, S. T., Van Der Heide, B., Langwell, L., & Walther, J. B. (2008). Too much of a good
thing? the relationship between number of friends and interpersonal impressions on
facebook. Journal of Computer‐Mediated Communication, 13(3), 531-549.
225
Tosun, L. P. (2012). Motives for Facebook use and expressing “true self” on the internet.
Computers in Human Behavior, 28(4), 1510-1517.
Trafimow, D., Triandis, H. C., & Goto, S. G. (1991). Some tests of the distinction between the
private self and the collective self. Journal of Personality and Social Psychology, 60(5),
649.
Triandis, H. C., Betancourt, H., Iwao, S., Leung, Kwok., Salazar, J. M., Setiadi, B., Sinha, J. B.
P., Touzard, H., & Zaleski, Z. (1993). An Etic-Emic Analysis of Individualism and
Collectivism. Journal of Cross-Cultural Psychology, 24(3), 366-383.
Triandis, H. C. (1989). The self and social behavior in differing cultural contexts. Psychological
Review, 96(3), 506.
Triandis, H. C., Bontempo, R., Villareal, M. J., Asai, M., & Lucca, N. (1988). Individualism and
collectivism: Cross-cultural perspectives on self-ingroup relationships. Journal of
Personality and Social Psychology, 54(2), 323.
Triandis, H. C., McCusker, C., & Hui, C. H. (1990). Multimethod probes of individualism and
collectivism. Journal of Personality and Social Psychology, 59(5), 1006.
Turner, J. C. (2005). Explaining the nature of power: A three‐process theory. European Journal
of Social Psychology, 35(1), 1-22.
Turner, J. C., & Brown, R. (1978). Social status, cognitive alternatives and intergroup relations.
Differentiation between Social Groups: Studies in the Social Psychology of Intergroup
Relations, , 201-234.
226
Turner, J. C., Hogg, M. A., Oakes, P. J., Reicher, S. D., & Wetherell, M. S. (1987).
Rediscovering the social group: A self-categorization theory. Basil Blackwell.
Turner, J. C., & Oakes, P. J. (1989). Self-categorization theory and social influence. Retrieved
July 5, 2016, from http://psycnet.apa.org/psycinfo/1989-97877-008
Turner, J. C., Wetherell, M. S., & Hogg, M. A. (1989). Referent informational influence and
group polarization. British Journal of Social Psychology, 28(2), 135-147.
Underwood, J. D., Kerlin, L., & Farrington-Flint, L. (2011). The lies we tell and what they say
about us: Using behavioural characteristics to explain facebook activity. Computers in
Human Behavior, 27(5), 1621-1626.
Valenzuela, S., Park, N., & Kee, K. F. (2009). Is there social capital in a social network site?:
Facebook use and college students' life satisfaction, trust, and participation1. Journal of
Computer‐Mediated Communication, 14(4), 875-901.
Valkenburg, P. M., Peter, J., & Schouten, A. P. (2006). Friend networking sites and their
relationship to adolescents' well-being and social self-esteem. CyberPsychology &
Behavior, 9(5), 584-590.
Van Widenfelt, B. M., Treffers, P. D., De Beurs, E., Siebelink, B. M., & Koudijs, E. (2005).
Translation and cross-cultural adaptation of assessment instruments used in psychological
research with children and families. Clinical Child and Family Psychology Review, 8(2),
135-147.
227
Viswanath, B., Mislove, A., Cha, M., & Gummadi, K. P. (2009). On the evolution of user
interaction in facebook. Proceedings of the 2nd ACM Workshop on Online Social Networks,
37-42.
Voci, A. (2006). The link between identification and in‐group favouritism: Effects of threat to
social identity and trust‐related emotions. British Journal of Social Psychology, 45(2), 265-
284.
Walczuch, R., & Lundgren, H. (2004). Psychological antecedents of institution-based consumer
trust in e-retailing. Information & Management, 42(1), 159-177.
Walther, J. B. (2007). Selective self-presentation in computer-mediated communication:
Hyperpersonal dimensions of technology, language, and cognition. Computers in Human
Behavior, 23(5), 2538-2557.
Walther, J. B., Van Der Heide, B., Hamel, L. M., & Shulman, H. C. (2009). Self-generated
versus other-generated statements and impressions in computer-mediated communication a
test of warranting theory using facebook. Communication Research, 36(2), 229-253.
Walther, J. B., Van Der Heide, B., Kim, S., Westerman, D., & Tong, S. T. (2008). The role of
friends’ appearance and behavior on evaluations of individuals on Facebook: Are we known
by the company we keep? Human Communication Research, 34(1), 28-49.
Wan, C., Chiu, C., Tam, K., Lee, S., Lau, I. Y., & Peng, S. (2007). Perceived cultural importance
and actual self-importance of values in cultural identification. Journal of Personality and
Social Psychology, 92(2), 337.
228
Wang, Q. (2004). The emergence of cultural self-constructs: Autobiographical memory and self-
description in European American and Chinese children. Developmental Psychology, 40(1),
3.
Wang, S. S., Moon, S., Kwon, K. H., Evans, C. A., & Stefanone, M. A. (2010). Face off:
Implications of visual cues on initiating friendship on Facebook. Computers in Human
Behavior, 26(2), 226-234.
Weisbuch, M., Ivcevic, Z., & Ambady, N. (2009). On being liked on the web and in the “real
world”: Consistency in first impressions across personal webpages and spontaneous
behavior. Journal of Experimental Social Psychology, 45(3), 573-576.
Wilson, J. M., Straus, S. G., & McEvily, B. (2006). All in due time: The development of trust in
computer-mediated and face-to-face teams. Organizational Behavior and Human Decision
Processes, 99(1), 16-33.
Wise, K., Alhabash, S., & Park, H. (2010). Emotional responses during social information
seeking on facebook. Cyberpsychology, Behavior, and Social Networking, 13(5), 555-562.
Yamagishi, T., & Kiyonari, T. (2000). The group as the container of generalized reciprocity.
Social Psychology Quarterly, , 116-132.
Yamaguchi, S., Greenwald, A. G., Banaji, M. R., Murakami, F., Chen, D., Shiomura, K., . . .
Krendl, A. (2007). Apparent universality of positive implicit self-esteem. Psychological
Science, 18(6), 498-500. doi:PSCI1928 [pii]
229
Zhang, S., Jiang, H., & Carroll, M. J. (2010). Social Identity in Faecbook Community Life.
International Journal of Virtual Communities and Social Networking, 2(4), 66-78.
Zhao, S., Grasmuck, S., & Martin, J. (2008). Identity construction on Facebook: Digital
empowerment in anchored relationships. Computers in Human Behavior, 24(5), 1816-1836.
Zimbardo, P. G. (1969). The human choice: Individuation, reason, and order versus
deindividuation, impulse, and chaos. Nebraska Symposium on Motivation,
Zywica, J., & Danowski, J. (2008). The faces of Facebookers: Investigating social enhancement
and social compensation hypotheses; predicting Facebook™ and offline popularity from
sociability and self‐esteem, and mapping the meanings of popularity with semantic
networks. Journal of Computer‐Mediated Communication, 14(1), 1-34.
230
Appendix 1
Study 1: Survey (Operationalization)
Facebook Culture
Thank you for volunteering to participate in this survey.
Before you undertake the survey please read the below information
carefully.
IMPORTANT -- Are you a member of Facebook?
If Yes -- Please read the information carefully and continue with the
survey.
If No -- Sorry you will not be able to participate in the study as this
study also looks at on line activities on Facebook.
231
Participation Information and Consent
Thank you for volunteering to participate in this online survey. Before you undertake the survey please
read the below information carefully.
Aims of the research:
This research looks at your activities and behaviour based on your interactions with your Facebook and non-Facebook (offline) contacts. This study is being undertaken for PhD research.
What will I be asked to do if I agree to take part?
All questions in the survey aim to understand your activities and behaviour based on your interactions with your Facebook and non-Facebook (offline) contacts. After each question you will either be asked to enter your response with a yes/no or a value or choose your best choice from a scale.
The first part is about the number and type of social contacts that you maintain and how you feel about
your social networks. In the second part you will be asked to describe how you are making use of Facebook. In addition, there are some questions about your general behaviour in social contexts. Please read the questions carefully before you respond as once submitted you cannot go back to the previous sections to amend your answers. After you make your choice please rate your response with your best answer. The survey will last approximately 20 - 30 minutes.
Will my data be anonymous?
Yes, your data will be anonymous. Name requirement is optional. However as this is a study on cultural
impact on communication behaviour your ethnicity, gender, age and country of residence, nationality
and first language will be required to make accurate inference from the data collected. Apart from this
no other personal information of identity will be required for this survey.
Summaries of non-personal data (data that can’t be linked to you) such as your responses and
demographic information may be retained long-term as part of a larger data set for publication or
teaching purposes.
Participation in the lottery is your decision for which your name and contact details will need to be
submitted. This information will be kept separate to the survey information and will only be used for
carrying out the lottery.
Do I have to take part?
Your participation in the survey will be entirely your choice. You are free to end the survey at any point
of your participation. If you want that your data be destroyed please contact me or my Director of
Studies within two weeks of completing the survey. My contact information is
[email protected] or my Director of Studies at [email protected] (see full contact
details below).
Important information to consider:
• This study has been ethically approved by the University Research Committee.
• Your responses will be kept anonymous and will be treated as strictly confidential. Responses
collated will be stored in a safe place once the survey is closed.
• The responses will be used for research purposes and the statistical analyses will be used in
conferences and journal articles.
232
If you have queries please contact:
Moon Moon Halder (PhD Researcher) Email: [email protected]
Director of studies: Dr. Jens Binder Email: [email protected] Nottingham Trent University School of Social Sciences Burton Street Nottingham
NG1 4BU
1) Declaration:
By continuing on the next page you agree that the information provided is enough for you to know
about the study and you agree to take part. This is also taken as confirmation that you are 18 or
above.
Please confirm by adding a tick (√) beside each sentence.
1) The study aims and designs has been clearly outlined.
2) I am aware that I can withdraw at any point of the study.
3) I am aware of what I would be expected to do in the study.
4) I am happy to volunteer.
233
2) Demographics:
Please tick (√) or give an answer where applicable:
a) Please add a unique code that will be used for you. This can be a combination of any number
or letter. This is the code that you can use to contact us in case you have any queries or if you
want to withdraw your data.
(e.g., your favourite colour, your favourite letter with any number, etc.)
b) Gender: Please tick (√) the your Gender : Male Female
c) Age: Please provide your Age in years:
d) Ethnicity: Please provide your Ethnicity (e.g., Indian, British, etc.)
e) Work: Please tick the one that is applicable to you.
i) Employed (full time) ii) Employed (part-time) iii) Unemployed
iv) Student (full time) v) Student (part-time)
f) Country of Residence: Please provide the name of the Country
g) Nationality: What is your Nationality?
h) Town/ City this survey is conducted in: Please provide the name of the Town or City
i) Language: What is your First Language?
j) What was your last English (subject) School/College/ University grade? (e.g., 40, 45,
65., etc.)
234
Your Social World:
3) Do you use any other Social Network sites apart from Facebook – if so, please select from the list.
a) Twitter b) WhatsApp c) Instagram d) LinkedIn
e) Bebo f) Orkut g) I don’t use any other social network site
4) Please provide the names of any other social network sites that you might also be using
(apart from the ones named above) in the box below (optional).
5) How often do you use other Social Network Sites apart from Facebook? (Please provide average
number of hours that you spend per day in numbers only).
6) Number of people I know on Facebook and Offline.
Offline contacts include members with who you interact in your day to day life and they might also be
on your Facebook contacts.
For this section you don’t have to count your social contacts on Facebook. Your response don’t have to
be accurate but a quick estimate of the number of people you know on Facebook. For the categories
you don’t know anyone please put a “0”.
Number of people I know on: FACEBOOK
OFFLINE
(this can also include contacts
who are also on your friends
list on Facebook)
Immediate family
235
Other birth family
Family of spouse or significant other
Co-workers
People at work but don't work with directly
Best friends/confidantes
People known through hobbies/recreation
People from religious organisations
People from other organisations
School relations
Neighbours
Just friends
People known through others
Childhood relations
People who provide a service
7) Think about your Facebook contacts now. Please give a quick estimate from how many of your
Facebook contacts you would seek advice, support or help in times of severe emotional or financial
crisis.
236
8) Again think about the people you know on Facebook. How often have you received the kind of
social support listed below.
None of the time
A little of the time
Some of the time
Most of the time
All of the time
i)Someone you can count on to listen to you when you need to talk
ii) Someone who provides you with information to help you understand a situation
iii) Someone to give you good advice
on crisis
iv) Someone to confide in or talk to about yourself or your problems
v) Someone whose advice you really want
vi) Someone to share your most private worries and fears with
vii) Someone whom you can turn to
deal with personal problems
viii) Someone who understands your problems
i) Someone with whom you can have a good time with
ii) Someone who can help you get your mind off things
Facebook Use:
9) Please indicate the number of days that you have used Facebook in the last 14 days.
10) Please provide the average number of hours (in minutes) that you have used Facebook in the last
14 days.
Additional Information: 1 hour = 60 minutes
So if you have spent an average of 2 hours in total on Facebook in the last 14 days, your answer should
be 120
237
11) For this part of the survey, please take a few minutes to think about your social interactions on
Facebook in the last 14 days. This indicates all situations where you have used your Facebook account.
Please note that situations here also includes interactions for which the other person was not present or
you have not received any response.
Please put an average number of minutes per day (in numbers only) spent on your social
interaction.
Additional Information: 1 hour = 60 minutes
So if you have spent an average of 1 hour in total on Facebook, your answer should be 60
Facebook used
Per weekday
(average number of
minutes spent)
Facebook used Per weekend day
(average number of minutes spent)
Study –used Facebook solely for
study purpose
Work - used Facebook solely for
work purpose
Social Life – used Facebook just for
socialising
Mixed - used Facebook for
work/study and socialising on at the
same time
Activities on Facebook:
12) In the last 14 days how often (on average) have you engaged in the following activities on
Facebook?
***Please rate the below activities based on your best ratings***
How often (on average) do you use Facebook for the following activities:
Never
Rarely Sometimes Often Very often
a) Playing Games
(one player games) b) Status Updates
c) Sharing Links
238
f) Private Messages
g) Commenting
h) Facebook Chat
i) Checking up
l) Events
m) Posting Photos
n) Tagging Photos
k) Viewing Photos
j) Posting Videos
d) Tagging Videos
e) Viewing Videos
13.a) In this section you will first be asked to rate the sentences on the left. These sentences are in
relation to Facebook use and your Facebook contacts. The next section (13.b) will have similar
questions but has to be answered in relation to your overall social contacts (Overall social contacts
includes your offline and Facebook contacts together).
***Please rate the following in relation to your Facebook contacts and Facebook use***
Disagree strongly
Disagree
Neither agree or disagree
Agree
Agree strongly
a) I have respect for elders with
whom I interact
b) It is important for me to
maintain harmony within my
social group
c) My happiness depends on the
happiness of those around me
d) I would offer my seat in a bus
to my professor
e) I respect people who are
humble about themselves
f) I will sacrifice my self-interest
for the benefit of the social
group I am in
g) I often have the feeling that my
relationships with others are more important than my own
accomplishments
239
h) I should take into consideration
my parents' advice when
making education/ career plans
i) It is important to me to respect
decisions made by my social
group
j) I will stay in a group if they
need me, even when I am not
happy with the group
k) If my brother or sister fails, I
feel responsible
l) Even when I strongly disagree
with my social group members,
I avoid an argument
13.b) In this section you will first be asked to rate the sentences on the left. These sentences are in
relation to your overall social contacts. (Overall social contacts includes your offline and Facebook
contacts together).
***Please rate the following in relation to your overall social contacts (Overall social contacts
includes your offline and Facebook contacts together).***
Please rate these below sentences based on your Overall Social Contacts
Disagree
strongly
Disagree
Neither
agree or
disagree
Agree
Agree
strongly
a) I have respect for elders with
whom I interact
b) It is important for me to
maintain harmony within my
social group
c) My happiness depends on the
happiness of those around me
d) I would offer my seat in a bus
to my professor
e) I respect people who are
humble about themselves
f) I will sacrifice my self-interest
for the benefit of the social
group I am in
g) I often have the feeling that my
relationships with others are
more important than my own
accomplishments
h) I should take into consideration
my parents' advice when
making education/ career plans
240
i) It is important to me to respect
decisions made by my social
group
j) I will stay in a group if they
need me, even when I am not
happy with the group
k) If my brother or sister fails, I
feel responsible
l) Even when I strongly disagree
with my social group members,
I avoid an argument
14.a) In this section you will first be asked to rate the sentences on the left. These sentences are in
relation to Facebook use and your Facebook contacts. The next section (14.b) will have similar
questions but has to be answered in relation to your Overall Social Contacts (Overall social contacts
includes your offline and Facebook contacts together).
***Please rate the following in relation to your Facebook contacts and Facebook use***
Disagree strongly
Disagree
Neither agree or disagree
Agree
Agree strongly
a) I'd rather say "No" directly,
than risk being misunderstood
b) Speaking up is not a problem
for me
c) Having a lively imagination is
important to me
d) I am comfortable on being
singled out for praise or
rewards
e) I am the same person at home
that I am during social
gathering
f) Being able to take care of
myself is a primary concern to
me
g) I act the same way no matter
who I am with
h) I feel comfortable using
someone's first name soon
after I meet them, even when
they are much older than I am
i) I prefer to be direct and
forthright when dealing with
people I've just met
241
j) I enjoy being unique and
different from others in many
respects
k) My personal identity is very
important to me
l) I value being in good health
above everything
14. b) In this section you will first be asked to rate the sentences on the left. These sentences are in
relation to your overall social contacts. (Overall social contacts includes your offline and Facebook
contacts together).
***Please rate the following in relation to your overall social contacts (Overall social contacts
includes your offline and Facebook contacts together).***
Disagree strongly
Disagree
Neither agree or disagree
Agree
Agree strongly
a) I'd rather say "No" directly,
than risk being misunderstood
b) Speaking up is not a problem
for me
c) Having a lively imagination is
important to me
d) I am comfortable on being
singled out for praise or
rewards
e) I am the same person at home
that I am during social
gathering
f) Being able to take care of
myself is a primary concern to
me
g) I act the same way no matter
who I am with
h) I feel comfortable using
someone's first name soon
after I meet them, even when
they are much older than I am
i) I prefer to be direct and
forthright when dealing with
people I've just met
j) I enjoy being unique and
different from others in many
respects
k) My personal identity is very
important to me
l) I value being in good health
above everything
242
Use of settings on Facebook:
15) This section looks at the use of settings on Facebook. Please give your truthful answers.
Please rate the below sentences as per your Facebook usage
Disagree strongly
Disagree
Neither agree or disagree
Agree
Agree strongly
a) I like to use filter setting to group my social contacts on
b) Using filter settings is important for me as it helps me to be open in my opinions on Facebook
c) I keep myself up-to-date with changes in privacy settings d) I don’t care who looks on my Facebook profile e) I always update my security
settings on my Facebook account f) I am fully aware of the use of privacy settings on Facebook g) I get worried about people
being able to view my personal information on Facebook
You and your social contacts:
16) In this section please rate yourself, your Facebook contacts and your Offline contacts
from a scale of 1 to 5.
Additional information: Facebook contacts includes your contacts that are on your Facebook contact
lists. It can also include Facebook members that you know offline as well.
Offline contacts includes your social contacts with whom you interact away from the online world but
might also be some of your Facebook contacts.
does not apply at all neutral applies very much
1 2 3 4 5
243
Attributes You 1 = does not apply at all 5 = applies very much
Contacts 1 = does not apply at all
5 = applies very much
Offline Social
Contacts 1 = does not apply at all
5 = applies very much
a) Respectful b) Independent c) Compliant d) Separate e) Tolerant f) Unconstrained g) Compromising h) Free i) Loyal j) Leader k) Self-sacrificing l) Unique m) Modest n) Original
17) This section looks at you and your social contacts on Facebook and how much you trust your
social contacts on Facebook. Please rate the below sentences truthfully.
Members on my
Facebook contacts:
Strongly
disagree
Disagree Neither agree
or disagree
Agree Strongly
agree
a) Do their best to help me
b) Do care about the well-being of others
c) Are open and receptive to the needs of each other
d) Are honest in dealing with each other
e) Keep their promises
f) Are trustworthy
244
18) This section is an attempt to check how much you identify with your social contacts on
Facebook. Please rate the below sentences truthfully.
Please rate the below sentences in relation to Facebook
Disagree strongly
Disagree
Neither agree or disagree
Agree
Agree strongly
a) I see myself as a member of my Facebook community.
b) It is important for me to be a member of my Facebook community.
c) I am not glad to be a member of my Facebook community.
d) I like being a member of my Facebook community.
e) I am not proud to be a member of my Facebook community.
f) I do not like being a member of my Facebook community.
You have done it. Thank you for your participation.
If you have any comments, complaints or suggestions, use the text box or contact the address
given below with your unique id and email/address (optional).
Contact address: Moon Moon Halder (PhD Researcher) Email: [email protected] Director of studies: Dr. Jens Binder
Comments/Suggestions:
245
Email: [email protected] Nottingham Trent University School of Social Sciences Burton Street Nottingham NG1 4BU
246
Debrief Form: Survey
Thank you for your participation! Aims of the Research:
This research looks at the impact of culture on communication behaviour and social identity on Online Social Media.
This study is being conducted in conjunction with an Experiment to investigate how members in a social
media environment can get primed due to environment factors around them which can have an impact
on their cultural and social identity. The data collected will be analysed using statistical methods. We
expect to find that priming will have an impact on culture and communication behaviour. Thank you for
taking part in the study. If you want to know more about this research, have any questions or
suggestions or simply want to find out the progress later please don’t hesitate to contact either me or
my Director of Studies.
Further, if you wish to withdraw your data from this study, please contact me within 14 days of the
study quoting your unique identifier.
If you have queries please contact:
Moon Moon Halder (PhD Researcher) Email: [email protected] Director of studies: Dr. Jens Binder Email: [email protected]
Nottingham Trent University
School of Social Sciences Burton Street Nottingham NG1 4BU UK
247
Appendix 2
Study 1: Survey (Facebook responses)
Facebook Culture
Thank you for volunteering to participate in this survey.
Before you undertake the survey please read the below information
carefully.
IMPORTANT -- Are you a member of Facebook?
If Yes -- Please read the information carefully and continue with the
survey.
If No -- Sorry you will not be able to participate in the study as this
study also looks at on line activities on Facebook.
248
Participation Information and Consent
Thank you for volunteering to participate in this online survey. Before you undertake the survey please
read the below information carefully.
Aims of the research:
This research looks at your activities and behaviour based on your interactions with your Facebook and non-Facebook (offline) contacts This study is being undertaken for PhD research.
What will I be asked to do if I agree to take part?
All questions in the survey aim to understand your activities and behaviour based on your interactions with your Facebook and non-Facebook (offline) contacts. After each question you will either be asked to enter your response with a yes/no or a value or choose your best choice from a scale.
The first part is about the number and type of social contacts that you maintain and how you feel about
your social networks. In the second part you will be asked to describe how you are making use of Facebook. In addition, there are some questions about your general behaviour in social contexts. Please read the questions carefully before you respond as once submitted you cannot go back to the previous sections to amend your answers. After you make your choice please rate your response with your best answer. The survey will last approximately 30 minutes
After the survey:
As a thank you for your support you can enter a lottery at the end of the survey. Each participant will
have the chance to win Amazon Vouchers worth £20.00. The winners will be informed by email.
Will my data be anonymous?
Yes, your data will be anonymous. Name requirement is optional. However as this is a study on cultural
impact on communication behaviour your ethnicity, gender, age and country of residence, nationality
and first language will be required to make accurate inference from the data collected. Apart from this
no other personal information of identity will be required for this survey.
Summaries of non-personal data (data that can’t be linked to you) such as your responses and
demographic information may be retained long-term as part of a larger data set for publication or
teaching purposes.
Participation in the lottery is your decision for which your name and contact details will need to be
submitted. This information will be kept separate to the survey information and will only be used for
carrying out the lottery.
Do I have to take part?
Your participation in the survey will be entirely your choice. You are free to end the survey at any point
of your participation. All you need to do is click the exit option. If you want that your data be destroyed
please contact me or my Director of Studies within two weeks of completing the survey. My contact
information is [email protected] or my Director of Studies at [email protected] (see
full contact details below).
Important information to consider:
• This study has been ethically approved by the University Research Committee.
• Your responses will be kept anonymous and will be treated as strictly confidential. Responses
collated will be stored in a safe place once the survey is closed.
249
• The responses will be used for research purposes and the statistical analyses will be used in
conferences and journal articles.
If you have queries please contact:
Moon Moon Halder (PhD Researcher) Email: [email protected] Director of studies: Dr. Jens Binder
Email: [email protected] Nottingham Trent University School of Social Sciences Burton Street
Nottingham
NG1 4BU
By continuing on the next page you agree that the information provided is enough for you to know
about the study and you agree to take part. This is also taken as confirmation that you are 18 or
above.
*****Advice – When filling the survey please try to give your truthful responses. Please
remember you will not be able to return back to your submitted answers so please take as
much time as possible on each question.*****
250
1) Demographics:
Please tick which is relevant to you:
a) Gender
Male Female
b) Age
Please provide your age in years
c)Ethnicity
How would you describe your ethnicity (e.g. White, British, British Asian, Indian, etc.)
d) Work
Employed (full time)
Employed (part time)
Unemployed Student (full time)
Student (part time)
e) Country of Residence
Please provide the name of the country
f) Nationality
Please provide your Nationality
g) Language
Your first Language
Your Social World:
2) Do you use any other Social Network sites apart from Facebook – if so, please select from the list.
a) Twitter b) Instagram c) LinkedIn d) Bebo e) Orkut
3) Please provide the names of any other social network sites that you might also be using
(apart from the ones named above) in the box below.
4) How often do you use other Social Network Sites apart from Facebook? Please provide average
number of hours that you spend per day in numbers only).
251
5) Number of people I know on Facebook and Online.
Number of people I know on: FACEBOOK
OFFLINE
(this can also include
contacts who are also on
your friends list on
Facebook)
Immediate family
Other birth family
Family of spouse or significant other
Co-workers
People at work but don't work with directly
Best friends/confidantes
People known through hobbies/recreation
People from religious organisations
People from other organisations
School relations
Neighbours
Just friends
People known through others
Childhood relations
252
People who provide a service
6) Think about your Facebook contacts now. Please give a quick estimate from how many of your
Facebook contacts you would seek advice, support or help in times of severe emotional or financial
crisis.
7) Again think about the people you know on Facebook. How often have you received the kind of
social support listed below.
None of
the time
A little of
the time
Some of
the time
Most of
the time
All of
the time
i)Someone you can count on to listen
to you when you need to talk ii) Someone who provides you with information to help you understand a situation
iii) Someone to give you good advice on crisis iv) Someone to confide in or talk to about yourself or your problems v) Someone whose advice you really want
vi) Someone to share your most private worries and fears with vii) Someone whom you can turn to deal with personal problems viii) Someone who understands your problems i) Someone with whom you can have a good time with ii) Someone who can help you get your mind off things
253
Facebook Use:
8) Please indicate the number of days that you have used Facebook in the last 14 days.
9) Please provide the average number of hours (in numbers) that you have used Facebook in the last
14 days.
10) For this part of the survey, please take a few minutes to think about your social interactions on
Facebook in the last 14 days. This indicates all situations where you have used your Facebook account.
Please note that situations here also includes interactions for which the other person was not present or
you have not received any response.
Please put an average number of hours per day (in numbers only) spent on your social
interaction.
Facebook used
Per weekday
Facebook used Per
weekend day
Study –used Facebook solely for study purpose
Work - used Facebook solely for work purpose
Social Life – used Facebook just for socialising
Mixed - used Facebook for work/study and
socialising on at the same time
Activities on Facebook:
11) In the last 14 days how often (on average) have you engaged in the following activities on
Facebook?
***Please rate the below activities based on your best ratings***
How often (on average) do you use Facebook for the following activities:
Never
Rarely Sometimes Often Very often
254
a) Playing Games
(one player games) b) Status Updates
c) Sharing Links
f) Private Messages
g) Commenting
h) Facebook Chat
i) Checking up
l) Events
m) Posting Photos
n) Tagging Photos
k) Viewing Photos
j) Posting Videos
d) Tagging Videos
e) Viewing Videos
12) In this section you will first be asked to rate some questions in relation to FACEBOOK. The next
section (11) will have similar questions but has to be answered in relation to your OVERALL SOCIAL
CONTACTS.
***Please rate the following in relation to your Facebook contacts***
Disagree strongly
Disagree
Neither agree or disagree
Agree
Agree strongly
a) I have respect for the authority
figures with whom I interact b) I’d rather say “No” directly, than
risk being misunderstood c) It is important for me to
maintain harmony within my group
d) Speaking up is not a problem for me
e) Having a lively imagination is important to me
f) I respect people who are modest about themselves
255
g) I will sacrifice my self-interest
for the benefit of the group I am in
h) Being able to take care of myself is a primary concern to me
i) I often have the feeling that my relationships with others are
more important than my own accomplishments
j) I act the same way no matter who I am with
k) It is important to me to respect
decisions made by the group l) I prefer to be direct and
forthright when dealing with people I’ve just met
m) I will stay in a group if they need me, even when I am not happy with the group
n) I enjoy being unique and different from others in many respects
o) My personal identity independent of others, is very important to me
p) Even when I strongly disagree
with group members, I avoid an argument
13) In this section please take into consideration your OVERALL SOCIAL CONTACTS
Please rate the following in relation to your Overall Social Contacts. This includes both Facebook and
Offline contacts.
Disagree strongly
Disagree
Neither agree or disagree
Agree
Agree strongly
a) I have respect for the authority figures with whom I interact
b) I’d rather say “No” directly, than
risk being misunderstood c) It is important for me to
maintain harmony within my group
d) Speaking up is not a problem for me
e) Having a lively imagination is important to me
f) I respect people who are modest about themselves
256
g) I will sacrifice my self-interest
for the benefit of the group I am in
h) Being able to take care of myself is a primary concern to me
i) I often have the feeling that my relationships with others are
more important than my own accomplishments
j) I act the same way no matter who I am with
k) It is important to me to respect
decisions made by the group l) I prefer to be direct and
forthright when dealing with people I’ve just met
m) I will stay in a group if they need me, even when I am not happy with the group
n) I enjoy being unique and different from others in many respects
o) My personal identity independent of others, is very important to me
p) Even when I strongly disagree
with group members, I avoid an argument
14) Use of settings on Facebook.
Please rate the following sentences as per your Facebook usage.
Disagree
strongly
Disagree
Neither
agree or disagree
Agree
Agree
strongly
I like to use filter setting to group my social contacts on Facebook Using filter settings is important for me as it helps me to be open in my opinions on Facebook
I keep myself up-to-date with changes in privacy settings I don’t care who looks on my Facebook profile I always update my security settings on my Facebook account I am fully aware of the use of privacy settings on Facebook
257
I get worried about people being
able to view my personal information on Facebook
You and your social contacts:
15) In this section please rate yourself as compared with your contacts.
Attributes You (0 = does not apply at all 5 = applies very much highest score)
Facebook Contacts (0 = does not apply at all 5 = applies very much highest score)
Offline Social Contacts (0 = does not apply at all 5 = applies very much highest score)
a) Respectful b) Independent c) Compliant d) Separate e) Tolerant f) Unconstrained g) Compromising h) Free i) Loyal j) Leader k) Self-sacrificing l) Unique m) Modest n) Original
16) In this section please rate the below sentences in relation to your social contacts connected to
your Facebook profile on how much you Trust your Facebook social contacts.
Members connected on my FACEBOOK PROFILE
will:
Strongly disagree
Disagree Neither agree or disagree
Agree Strongly agree
a) Do their best to help me
b) Do care about the well-being of others
c) Are open and receptive to the needs of each other
258
d) Are honest in dealing with each other
e) Keep their promises
f) Are trustworthy
17) In this section please rate yourself in comparison with your social contacts on your friends
list on Facebook.
Disagree strongly
Disagree
Neither agree or
disagree
Agree
Agree strongly
a) I see myself as a member of my Facebook community.
b) It is important for me to be a member of my Facebook community.
c) I am not glad to be a member of my Facebook community.
d) I like being a member of my Facebook community.
e) I am not proud to be a member of my Facebook community.
f) I do not like being a member of my Facebook community.
259
You have done it. Thank you for your participation.
If you want to be entered in the lottery draw to have a chance to win Amazon Vouchers worth
£20.00, please enter your name, contact details and email address below. Please be assured
these details will solely be used for lottery purpose and will be destroyed once the draw has
been completed.
If you have any comments, complaints or suggestions, use the text box or contact the address
given below.
Contact address: Moon Moon Halder (PhD Researcher) Email: [email protected] Director of studies: Dr. Jens Binder Email: [email protected] Nottingham Trent University School of Social Sciences Burton Street Nottingham NG1 4BU
Name:
Address:
Email address:
Comments/Suggestions:
260
Debrief Form: Survey
Thank you for your participation! Aims of the Research: This research looks at the impact of culture on communication behaviour and social identity on Online
Social Media.
This study is being conducted in conjunction with an Experiment to investigate how members in a social
media environment can get primed due to environment factors around them which can have an impact
on their cultural and social identity. The data collected will be analysed using statistical methods. We
expect to find that priming will have an impact on culture and communication behaviour. Thank you for
taking part in the study. If you want to know more about this research, have any questions or
suggestions or simply want to find out the progress later please don’t hesitate to contact either me or
my Director of Studies.
Further, if you wish to withdraw your data from this study, please contact me within 14 days of the
study quoting your unique identifier.
If you have queries please contact:
Moon Moon Halder (PhD Researcher)
Email: [email protected] Director of studies: Dr. Jens Binder Email: [email protected]
Nottingham Trent University School of Social Sciences
Burton Street Nottingham NG1 4BU UK
261
Appendix 3
Study 3: Experiment– collectivism version
Circulated in India
Participation Information and Consent
Thank you for volunteering to participate in this survey. Before you undertake the survey
please read the below information carefully.
IMPORTANT – Are you a member of Facebook?
Yes – Please read the information below and participate in the study.
No – Sorry you will not be able to participate in the study as this study also looks at online
activities.
Aims of the research:
This research looks at your activities and behaviour based on your interactions with your Facebook and non-Facebook (offline) contacts. This study is being undertaken for PhD research.
What will I be asked to do if I agree to take part?
All questions in the survey aim to understand your activities and behaviour based on your interactions with your Facebook and non-Facebook (offline) contacts. After each question you will either be asked to enter your response with a yes/no or a value or choose your best choice from a scale.
The first part is about the number and type of social contacts that you maintain and how you feel about your social networks. In the second part you will be asked to describe how you are making use of Facebook. In addition, there are some questions about your general behaviour in social contexts.
Please read the questions carefully before you respond as once submitted you cannot go back to the previous sections to amend your answers. After you make your choice please rate your response with your best answer. The survey will last approximately 15-20 minutes
After the survey:
262
As a thank you for your support you can enter a lottery at the end of the survey. Each participant
will have the chance to win voucher for Rs.500 which will be awarded to two lucky winners. The
winners will be informed by email.
Will my data be anonymous?
Yes, you data will be anonymous. Name requirement is optional. However as this is a study on
cultural impact on communication behaviour your ethnicity, gender, age and country of
residence, nationality and first language will be required to make accurate inference from the
data collected. Apart from this no other personal information of identity will be required for this
survey.
Summaries of non-personal data (data that can’t be linked to you) such as your responses and
demographic information may be retained long-term as part of a larger data set for publication
or teaching purposes.
Your data will not be used on its own in isolation; instead your data will be analyzed alongside
the data of all other respondents and only general trends and patterns will be reported and your
confidentiality and anonymity will be protected at all times.
Participation in the lottery is your decision for which your name and contact details will need to
be submitted. This information will be kept separate to the survey information and will only be
used for carrying out the lottery.
Do I have to take part?
Your participation in the survey will be entirely your choice. There are no foreseen negative
consequences of taking part in this research.
You are free to end the survey at any point of your participation. All you need to do is click the
exit option. If you want that your data be destroyed, please contact me or my Director of Studies
within two weeks of completing the survey.
Important information to consider:
• This study has been ethically approved by the University Research Committee.
• Your responses will be kept anonymous and will be treated as strictly confidential.
Responses collated will be stored in a safe place once the survey is closed.
• The responses will be used for research purposes and the statistical analyses will be
used in conferences and journal articles.
If you have queries or suggestions please contact:
263
PhD Researcher: Moon Moon Halder (Email: [email protected]) Director of studies: Dr. Jens Binder (Email: [email protected]) Nottingham Trent University School of Social Sciences Burton Street Nottingham NG1 4BU
Please add “X” in the box below to confirm your agreement to participate in the
study:
1) The study aims and design has been clearly outlined-
2) I can withdraw at any point of the study -
3) I am aware of what I would be expected to do in the study -
4) I am happy to volunteer –
264
1) Please fill in the boxes:
a) Please enter your Unique Reference Number (this can be used later if you want to refer back
at your data)
b) Gender
(Please place a ‘X’ under your choice)
Male Female
c) Age
Please provide your age in years
d) Ethnicity
How would you describe your ethnicity (e.g. White, British, British Asian, Indian, etc.)
e) Work
Employed (full time)
Employed (part time)
Unemployed Student (full time)
Student (part time)
f) Country of Residence
Please provide the name of the country
g) Nationality
Please provide your Nationality
h) English School Grade
What was your English School Grade in the last
term or last year?
265
2) In this section please rate the following based on your best
choice. Please add an “X” for your response.
Disagree strongly
Disagree
Neither agree or disagree
Agree
Agree strongly
a) I have respect for the
authority
figures with whom I interact
b) I’d rather say “No” directly,
than risk being misunderstood
c) It is important for me to
maintain harmony within my
group
d) Speaking up is not a problem
for me -
e) My happiness depends on the
happiness of those around me
f) Having a lively imagination is
important to me
g) I would offer my seat in a bus
to my professor
h) I am comfortable on being
singled out for praise or
rewards
i) I respect people who are
modest about themselves
j) I am the same person at
home that I am during social
gathering
k) I will sacrifice my self-interest
for the benefit of the group I
am in
l) Being able to take care of
myself is a primary concern to
me
m) I often have the feeling that
my relationships with others
are more important than my
own accomplishments
n) I act the same way no matter
who I am with
o) I should take into
consideration my parents’
advice when making
education/ career plans
p) I feel comfortable using
someone’s first name soon
266
after I meet them, even when
they are much older than I am
q) It is important to me to
respect decisions made by the
group
r) I prefer to be direct and
forthright when dealing with
people I’ve just met
s) I will stay in a group if they
need me, even when I am not
happy with the group
t) I enjoy being unique and
different from others in many
respects
u) If my brother or sister fails, I
feel responsible
v) My personal identity
independent of others, is very
important to me
w) Even when I strongly disagree
with group members, I avoid
an argument
x) I value being in good health
above everything
267
3) Which category of people do you best fit in? Please tick one of
the category that best reflects you.
a) I value individual achievement and strongly believe in my personal
goals. My personal achievements and success is of priority for me. I
strongly believe in my own efforts. I value independence.
1) Not at all
2) A little bit
3) Not sure
4) Quite a lot
5) Very much
a) I value the importance of my relationship with others. My goals are
achievable/ achieved with the help of and support of others around me.
Living a life of harmony with others is of priority for me.
1) Not at all
2) A little bit
3) Not sure
4) Quite a lot
5) Very much
268
4) Please read this passage carefully and provide your truthful
response.
Sostoras, a warrior in ancient Sumer, was largely responsible for the success of
Sargon I in conquering all of Mesopotamia. As a result, he was rewarded with a
small kingdom of his own to rule.
About 10 years later, Sargon I was conscripting warriors for a war. Sostoras was
obligated to send a detachment of soldiers to aid Sargon I. He had to decide who
to put in command of the detachment. After thinking about it for a long time,
Sostoras eventually decided on Tiglath who was a member of his family.
This appointment had several advantages. Sostoras was able to show his loyalty to
his family. He was also able to cement their loyalty to him. In addition, having
Tiglath as the commander increased the power and prestige of the family.
Answer the question:
Do you think Sostoras was a good choice? Circle the appropriate answer.
Yes
No
Not sure
269
5) As a next task concentrate on the text below. Please circle all pronouns (such
as we, us, and our) that appear in it.
We go to the park. Our excitement fills us when we see the ice-cream van. We allow
ourselves to explore every corner of the park, never letting other people distract
us. Our voice fills the air and street. We window shop and everywhere we go we
see our reflection looking back at us in the glass from the shops we walk past. When
we return home, our hearts fill with joy and happiness as we know that we will soon
return back in the park. The park belongs to us.
6) The next task is a short writing exercise.
For the next few minutes please read the questions below and answer
the following.
a) List 4 things that you would like to obtain for yourself to improve your
everyday life.
i)…………………….
ii)……………………
iii)…………………..
iv)…………………..
b) List 4 things that you value about yourself as a person.
i)…………………….
ii)……………………
iii)…………………..
iv)…………………..
270
a) Please think of what you have in common with your family and friends
b) What is your goal for the next 1 year in relation to your family and
friends?
Well done! You have done a great job, just few more to go.
271
7) There are twenty number blanks on the page below. Please write twenty
answers to the simple question ‘Who am I?’ in the blanks. Just give twenty
different answers to this question. Answer as if you were giving the answers
to yourself, not to somebody else. Write the answers in order that they occur
to you. Don’t worry about the logic or ‘importance’. Go along fairly fast as time
is limited.
1) I am …………………………………………………..
2) I am …………………………………………………..
3) I am …………………………………………………..
4) I am …………………………………………………..
5) I am …………………………………………………..
6) I am …………………………………………………..
7) I am …………………………………………………..
8) I am …………………………………………………..
9) I am …………………………………………………..
10) I am …………………………………………………..
272
8) Please have a think about your Facebook usage and your Facebook
social group and rate the following items/ questions truthfully. Please
select only one answer to each question. Please tick (√) or circle one
response per question.
i) Privacy settings:
a) How confident are you that the information that you upload on Facebook are not
misused by others?
1 - Not at all
2 – A little bit
3 – Not sure
4 – Quite a lot
5 - Very Much
b) How safe do you feel when you are on Facebook?
1 - Not at all
2 – A little bit
3 – Not sure
4 – Quite a lot
5 - Very Much
c) While on Facebook do you feel you are at a particular risk?
1 - Not at all
2 – A little bit
3 – Not sure
4 – Quite a lot
5 - Very Much
d) Do you feel that others are at risk because of using Facebook?
1 - Not at all
2 – A little bit
3 – Not sure
4 – Quite a lot
5 - Very Much
ii) Self-enhancement
a) When I use Facebook I feel less constrained. It makes me feel free.
273
1 - Not at all
2 – A little bit
3 – Not sure
4 – Quite a lot
5 - Very Much
b) I feel that I am better than my friends on Facebook. Using Facebook gives me a sense
of self-worth.
1 - Not at all
2 – A little bit
3 – Not sure
4 – Quite a lot
5 - Very Much
c) I am loyal to my social community on Facebook. My social community means a lot to
me in my daily life.
1 - Not at all
2 – A little bit
3 – Not sure
4 – Quite a lot
5 - Very Much
d) I believe that having a highly flattering profile is important to me as it attracts people’s
affection and admiration for me.
1 - Not at all
2 – A little bit
3 – Not sure
4 – Quite a lot
5 - Very Much
e) I like to be unique on Facebook. Hence I like posting my achievements for sharing it
with others.
1 - Not at all
2 – A little bit
3 – Not sure
4 – Quite a lot
5 - Very Much
f) I am often tolerant to others on Facebook. I don’t feel it is necessary to engage in
arguments with my social community.
1 - Not at all
2 – A little bit
3 – Not sure
274
4 – Quite a lot
5 - Very Much
iii) Social Support:
a) How likely would you offer social support to others?
1 - Not at all
2 – A little bit
3 – Not sure
4 – Quite a lot
5 - Very Much
b) How likely do you feel that you are connected to your Facebook community?
1 - Not at all
2 – A little bit
3 – Not sure
4 – Quite a lot
5 - Very Much
c) How likely do you feel that members of your social group would help you during a
personal crisis?
1 - Not at all
2 – A little bit
3 – Not sure
4 – Quite a lot
5 - Very Much
d) How likely are you to involve yourself with a social cause on Facebook?
1 - Not at all
2 – A little bit
3 – Not sure
4 – Quite a lot
5 - Very Much
iv) Self-identification:
a) I feel strongly connected with my group members on Facebook
1 – Never
2 – Seldom
3 – Sometimes
275
4 – Often
5 – Very Often
b) I often feel held back by my group on Facebook
1 – Never
2 – Seldom
3 – Sometimes
4 – Often
5 – Very Often
c) My Facebook friends are very important for me.
1 – Never
2 – Seldom
3 – Sometimes
4 – Often
5 – Very Often
d) I sometimes makes excuses of belonging to my friends group on Facebook
1 – Never
2 – Seldom
3 – Sometimes
4 – Often
5 – Very Often
v) Trust:
a) My Facebook community are trustworthy
1 – Never
2 – Seldom
3 – Sometimes
4 – Often
5 – Very Often
b) I am confident that my friends on Facebook mostly shares their honest opinions
about me
1 – Never
2 – Seldom
3 – Sometimes
4 – Often
5 – Very Often
c) I trust Facebook for safeguarding my personal information
276
1 – Never
2 – Seldom
3 – Sometimes
4 – Often
5 – Very Often
d) My Facebook community does their best to help me whenever I need them.
1 – Never
2 – Seldom
3 – Sometimes
4 – Often
5 – Very Often
9) How likely are you to act in a particular way?
Scenario 1:
I log on my Facebook and I receive a friend request from my family member. How likely is
that you will accept the friend request?
Please select only one answer
1) Very un-likely
2) Not so likely
3) Not sure
4) Quite likely
5) Very likely
Scenario 2:
I log on my Facebook and receive a friend request from an unknown person. How likely will
you will accept the friend request?
Please select only one answer
1) Very un-likely
2) Not so likely
3) Not sure
4) Quite likely
5) Very likely
Scenario 3:
277
While logging on Facebook I am reminded to update my privacy settings. How likely would
you update your privacy settings.
Please select only one answer
1) Very un-likely
2) Not so likely
3) Not sure
4) Quite likely
5) Very likely
Scenario 4:
While on Facebook I get a request to join a protest for a noble/ social cause. How likely would
you join the protest.
Please select only one answer
1) Very un-likely
2) Not so likely
3) Not sure
4) Quite likely
5) Very likely
Scenario 5:
While on Facebook when I see my friends talk about their achievements. I would congratulate
them.
Please select only one answer
1) Very un-likely
2) Not so likely
3) Not sure
4) Quite likely
5) Very likely
Scenario 6:
While on Facebook I often seem to join social groups that has personal relevance to me.
Please select only one answer
1) Very un-likely
2) Not so likely
3) Not sure
4) Quite likely
5) Very likely
278
10) Please fill out this short questionnaire and provide truthful answers as it
will help me to spot areas of improvement in my study.
1) Do you think this study put a focus on –
(Please tick (√) the correct option)
a) Social Obligation
1 - Not at all
2 - A little bit
3 - Somewhat
4 - Not sure/ N/A
5 - Quite a bit
6 - Quite a lot
7 - Very Much
b) Individual decisions
1 - Not at all
2 - A little bit
3 - Somewhat
4 - Not sure/ N/A
5 - Quite a bit
6 - Quite a lot
7 - Very Much
c) Reading skills
1 - Not at all
2 - A little bit
3 - Somewhat
4 - Not sure/ N/A
5 - Quite a bit
6 - Quite a lot
7 - Very Much
279
……………………………………………………………………………………………………………………
11) How did you find the instructions? Was it easy to follow or do you think it
can be improved?
9) Please provide any suggestions you think could help my study to
improve.
You have done it. Thank you for your participation.
If you want to be entered in the lottery draw to have a chance to win a gift
voucher worth Rs. 500, please enter your name, contact details and email
address below. Two lucky winners will be selected from the lucky draw. Please
be assured these details will solely be used for lottery purpose and will be
destroyed once the draw has been completed.
If you have any comments, complaints or suggestions, use the text box
contact the address given below.
Name:
Address:
Email address:
280
Contact address:
Moon MoonHalder (PhD Researcher)
Email: [email protected]
Director of studies: Dr. Jens Binder
Email: [email protected]
Nottingham Trent University School of Social Sciences
Burton Street Nottingham
NG1 4BU
281
Debrief Form:
Thank you for your participation! Aims of the Research:
This is a part of a wider project that looks at the impact of culture on communication behaviour
and social identity on Online Social Media.
This study investigates how individuals can get primed due to environment factors around
them which can have an impact on their cultural and social identity. We expect to find that
priming will have an impact on culture and, ultimately, on communication behaviour. Thank
you for taking part in the study. If you want to know more about this research, have any
questions or suggestions or simply want to find out the progress later please don’t hesitate
to contact either me or my Director of Studies.
Further, if you wish to withdraw your data from this study, please contact me within 14
days of the study quoting your unique identifier.
If you have queries please contact:
Moon MoonHalder (PhD Researcher)
Email: [email protected]
Director of studies: Dr. Jens Binder
Email: [email protected]
Nottingham Trent University
School of Social Sciences
Burton Street
Nottingham
NG1 4BU UK
282
Appendix 4
Study 3: Experiment: Individualism version
Used in India
Participation Information and Consent
Thank you for volunteering to participate in this survey. Before you undertake the survey
please read the below information carefully.
IMPORTANT – Are you a member of Facebook?
Yes – Please read the information below and participate in the study.
No – Sorry you will not be able to participate in the study as this study also looks at online
activities.
Aims of the research:
This research looks at your activities and behaviour based on your interactions with your Facebook and non-Facebook (offline) contacts. This study is being undertaken for PhD research.
What will I be asked to do if I agree to take part?
All questions in the survey aim to understand your activities and behaviour based on your interactions with your Facebook and non-Facebook (offline) contacts. After each question you will either be asked to enter your response with a yes/no or a value or choose your best choice from a scale.
The first part is about the number and type of social contacts that you maintain and how you feel about your social networks. In the second part you will be asked to describe how you are making use of Facebook. In addition, there are some questions about your general behaviour in social contexts.
Please read the questions carefully before you respond as once submitted you cannot go back to the previous sections to amend your answers. After you make your choice please rate your response with your best answer. The survey will last approximately 15-20 minutes
After the survey:
283
As a thank you for your support you can enter a lottery at the end of the survey. Each participant
will have the chance to win voucher for Rs.500 which will be awarded to two lucky winners. The
winners will be informed by email.
Will my data be anonymous?
Yes, you data will be anonymous. Name requirement is optional. However as this is a study on
cultural impact on communication behaviour your ethnicity, gender, age and country of
residence, nationality and first language will be required to make accurate inference from the
data collected. Apart from this no other personal information of identity will be required for this
survey.
Summaries of non-personal data (data that can’t be linked to you) such as your responses and
demographic information may be retained long-term as part of a larger data set for publication
or teaching purposes.
Your data will not be used on its own in isolation; instead your data will be analyzed alongside
the data of all other respondents and only general trends and patterns will be reported and your
confidentiality and anonymity will be protected at all times.
Participation in the lottery is your decision for which your name and contact details will need to
be submitted. This information will be kept separate to the survey information and will only be
used for carrying out the lottery.
Do I have to take part?
Your participation in the survey will be entirely your choice. There are no foreseen negative
consequences of taking part in this research.
You are free to end the survey at any point of your participation. All you need to do is click the
exit option. If you want that your data be destroyed please contact me or my Director of Studies
within two weeks of completing the survey.
Important information to consider:
• This study has been ethically approved by the University Research Committee.
• Your responses will be kept anonymous and will be treated as strictly confidential.
Responses collated will be stored in a safe place once the survey is closed.
• The responses will be used for research purposes and the statistical analyses will be
used in conferences and journal articles.
If you have queries or suggestions please contact:
284
PhD Researcher: Moon Moon Halder (Email: [email protected]) Director of studies: Dr. Jens Binder (Email: [email protected]) Nottingham Trent University School of Social Sciences Burton Street Nottingham NG1 4BU
Please add “X” in the box below to confirm your agreement to participate in the
study:
5) The study aims and design has been clearly outlined-
6) I can withdraw at any point of the study -
7) I am aware of what I would be expected to do in the study -
8) I am happy to volunteer –
285
2) Please fill in the boxes:
c) Please enter your Unique Reference Number (this can be used later if you want to refer back
at your data)
d) Gender
(Please place a ‘X’ under your choice)
Male Female
c) Age
Please provide your age in years
e) Ethnicity
How would you describe your ethnicity (e.g. White, British, British Asian, Indian, etc.)
e) Work
Employed (full time)
Employed (part time)
Unemployed Student (full time)
Student (part time)
f) Country of Residence
Please provide the name of the country
g) Nationality
Please provide your Nationality
h) English School Grade
What was your English School Grade in the last
term or last year?
286
2) In this section please rate the following based on your best
choice. Please add an “X” for your response.
Disagree strongly
Disagree
Neither agree or disagree
Agree
Agree strongly
a) I have respect for the
authority
figures with whom I interact
b) I’d rather say “No” directly,
than risk being misunderstood
c) It is important for me to
maintain harmony within my
group
d) Speaking up during a class is
not a problem for me -
e) My happiness depends on the
happiness of those around me
f) Having a lively imagination is
important to me
g) I would offer my seat in a bus
to my professor
h) I am comfortable with being
singled out for praise or
rewards
i) I respect people who are
modest about themselves
j) I am the same person at
home that I am during social
gathering
k) I will sacrifice my self-interest
for the benefit of the group I
am in
l) Being able to take care of
myself is a primary concern to
me
m) I often have the feeling that
my relationships with others
are more important than my
own accomplishments
n) I act the same way no matter
who I am with
o) I should take into
consideration my parents’
advice when making
education/ career plans
p) I feel comfortable using
someone’s first name soon
287
after I meet them, even when
they are much older than I am
q) It is important to me to
respect decisions made by the
group
r) I prefer to be direct and
forthright when dealing with
people I’ve just met
s) I will stay in a group if they
need me, even when I am not
happy with the group
t) I enjoy being unique and
different from others in many
respects
u) If my brother or sister fails, I
feel responsible
v) My personal identity
independent of others, is very
important to me
w) Even when I strongly disagree
with group members, I avoid
an argument
x) I value being in good health
above everything
288
4) Which category of people do you best fit in? Please tick one of
the category that best reflects you.
b) I value individual achievement and strongly believe in my personal
goals. My personal achievements and success is of priority for me. I
strongly believe in my own efforts. I value independence.
1) Not at all
2) A little bit
3) Not sure
4) Quite a lot
5) Very much
b) I value the importance of my relationship with others. My goals are
achievable/ achieved with the help of and support of others around me.
Living a life of harmony with others is of priority for me.
1) Not at all
2) A little bit
3) Not sure
4) Quite a lot
5) Very much
289
4) Please read this passage carefully and provide your truthful
response.
Sostoras, a warrior in ancient Sumer, was largely responsible for the success
of Sargon 1 in conquering all of Mesopotamia. As a result, he was rewarded
with a small kingdom of his own to rule.
About 10 years later, Sargon 1 was conscripting warriors for a war. Sostoras
was obligated to send a detachment of soldiers to aidSargon 1. He had to
decide who to put in command of the detachment.After thinking about it for a
long time, Sostoras eventually decided onTiglath who was a talented general.
This appointment had several advantages. Sostoras was able to makean
excellent general indebted to him. This would solidify Sostoras'shold on his
own dominion. In addition, the very fact of having a general such as Tiglath
as his personal representative would greatly increase Sostoras's prestige.
Finally, sending his best general would be likely to make Sargon 1 grateful.
Consequently, there was the possibility of getting rewarded by Sargon 1.
Answer the question:
Do you admire Sostoras? Circle the appropriate answer.
Yes
No
Not sure
290
5) As a next task concentrate on the text below. Please circle all pronouns (such
as I, me, and my) that appear in it.
I go to the park. My excitement fills me when I see the ice-cream van. I allow myself
to explore every corner of the park, never letting other people distract me. My
voice fills the air and street. I window shop and everywhere I go I see my reflection
looking back at me in the glass from the shops I walk past. When I return home, my
hearts fills with joy and happiness as I know that I will soon return back in the park.
The park belongs to me.
6) The next task is a short writing exercise.
For the next few minutes please read the questions below and answer
the following.
c) List 4 things that you would like to obtain for yourself to improve your
everyday life.
i)…………………….
ii)……………………
iii)…………………..
iv)…………………..
d) List 4 things that you value about yourself as a person.
i)…………………….
ii)……………………
iii)…………………..
iv)…………………..
291
e) Please think of what makes you different from your family and friends
f) What is your personal goal for the next 1 year?
292
Well done! You have done a great job, just few more to go.
7) There are twenty number blanks on the page below. Please write twenty
answers to the simple question ‘Who am I?’ in the blanks. Just give twenty
different answers to this question. Answer as if you were giving the answers
to yourself, not to somebody else. Write the answers in order that they occur
to you. Don’t worry about the logic or ‘importance’. Go along fairly fast as time
is limited.
11) I am …………………………………………………..
12) I am …………………………………………………..
13) I am …………………………………………………..
14) I am …………………………………………………..
15) I am …………………………………………………..
16) I am …………………………………………………..
17) I am …………………………………………………..
18) I am …………………………………………………..
19) I am …………………………………………………..
20) I am …………………………………………………..
293
8) Please have a think about your Facebook usage and your Facebook
social group and rate the following items/ questions truthfully. Please
select only one answer to each question. Please tick (√) or circle one
response per question.
vi) Privacy settings:
e) How confident are you that the information that you upload on Facebook are not
misused by others?
1 - Not at all
2 – A little bit
3 – Not sure
4 – Quite a lot
5 - Very Much
f) How safe do you feel when you are on Facebook?
1 - Not at all
2 – A little bit
3 – Not sure
4 – Quite a lot
5 - Very Much
g) While on Facebook do you feel you are at a particular risk?
1 - Not at all
2 – A little bit
3 – Not sure
4 – Quite a lot
5 - Very Much
h) Do you feel that others are at risk because of using Facebook?
1 - Not at all
2 – A little bit
3 – Not sure
4 – Quite a lot
5 - Very Much
vii) Self-enhancement
g) When I use Facebook I feel less constrained. It makes me feel free.
294
1 - Not at all
2 – A little bit
3 – Not sure
4 – Quite a lot
5 - Very Much
h) I feel that I am better than my friends on Facebook. Using Facebook gives me a sense
of self-worth.
1 - Not at all
2 – A little bit
3 – Not sure
4 – Quite a lot
5 - Very Much
i) I am loyal to my social community on Facebook. My social community means a lot to
me in my daily life.
1 - Not at all
2 – A little bit
3 – Not sure
4 – Quite a lot
5 - Very Much
j) I believe that having a highly flattering profile is important to me as it attracts people’s
affection and admiration for me.
1 - Not at all
2 – A little bit
3 – Not sure
4 – Quite a lot
5 - Very Much
k) I like to be unique on Facebook. Hence I like posting my achievements for sharing it
with others.
1 - Not at all
2 – A little bit
3 – Not sure
4 – Quite a lot
5 - Very Much
l) I am often tolerant to others on Facebook. I don’t feel it is necessary to engage in
arguments with my social community.
1 - Not at all
2 – A little bit
3 – Not sure
295
4 – Quite a lot
5 - Very Much
viii) Social Support:
e) How likely would you offer social support to others?
1 - Not at all
2 – A little bit
3 – Not sure
4 – Quite a lot
5 - Very Much
f) How likely do you feel that you are connected to your Facebook community?
1 - Not at all
2 – A little bit
3 – Not sure
4 – Quite a lot
5 - Very Much
g) How likely do you feel that members of your social group would help you during a
personal crisis?
1 - Not at all
2 – A little bit
3 – Not sure
4 – Quite a lot
5 - Very Much
h) How likely are you to involve yourself with a social cause on Facebook?
1 - Not at all
2 – A little bit
3 – Not sure
4 – Quite a lot
5 - Very Much
ix) Self-identification:
e) I feel strongly connected with my group members on Facebook
1 – Never
2 – Seldom
3 – Sometimes
296
4 – Often
5 – Very Often
f) I often feel held back by my group on Facebook
1 – Never
2 – Seldom
3 – Sometimes
4 – Often
5 – Very Often
g) My Facebook friends are very important for me.
1 – Never
2 – Seldom
3 – Sometimes
4 – Often
5 – Very Often
h) I sometimes makes excuses of belonging to my friends group on Facebook
1 – Never
2 – Seldom
3 – Sometimes
4 – Often
5 – Very Often
x) Trust:
e) My Facebook community are trustworthy
1 – Never
2 – Seldom
3 – Sometimes
4 – Often
5 – Very Often
f) I am confident that my friends on Facebook mostly shares their honest opinions
about me
1 – Never
2 – Seldom
3 – Sometimes
4 – Often
5 – Very Often
g) I trust Facebook for safeguarding my personal information
297
1 – Never
2 – Seldom
3 – Sometimes
4 – Often
5 – Very Often
h) My Facebook community does their best to help me whenever I need them.
1 – Never
2 – Seldom
3 – Sometimes
4 – Often
5 – Very Often
9) How likely are you to act in a particular way?
Scenario 1:
I log on my Facebook and I receive a friend request from my family member. How likely is
that you will accept the friend request?
Please select only one answer
6) Very un-likely
7) Not so likely
8) Not sure
9) Quite likely
10) Very likely
Scenario 2:
I log on my Facebook and receive a friend request from an unknown person. How likely will
you will accept the friend request?
Please select only one answer
6) Very un-likely
7) Not so likely
8) Not sure
9) Quite likely
10) Very likely
Scenario 3:
298
While logging on Facebook I am reminded to update my privacy settings. How likely would
you update your privacy settings.
Please select only one answer
6) Very un-likely
7) Not so likely
8) Not sure
9) Quite likely
10) Very likely
Scenario 4:
While on Facebook I get a request to join a protest for a noble/ social cause. How likely would
you join the protest.
Please select only one answer
6) Very un-likely
7) Not so likely
8) Not sure
9) Quite likely
10) Very likely
Scenario 5:
While on Facebook when I see my friends talk about their achievements. I would congratulate
them.
Please select only one answer
6) Very un-likely
7) Not so likely
8) Not sure
9) Quite likely
10) Very likely
Scenario 6:
While on Facebook I often seem to join social groups that has personal relevance to me.
Please select only one answer
6) Very un-likely
7) Not so likely
8) Not sure
9) Quite likely
10) Very likely
299
10) Please fill out this short questionnaire and provide truthful answers as it
will help me to spot areas of improvement in my study.
2) Do you think this study put a focus on –
(Please tick (√) the correct option)
d) Social Obligation
1 - Not at all
2 - A little bit
3 - Somewhat
4 - Not sure/ N/A
5 - Quite a bit
6 - Quite a lot
7 - Very Much
e) Individual decisions
1 - Not at all
2 - A little bit
3 - Somewhat
4 - Not sure/ N/A
5 - Quite a bit
6 - Quite a lot
7 - Very Much
f) Reading skills
1 - Not at all
2 - A little bit
3 - Somewhat
4 - Not sure/ N/A
5 - Quite a bit
6 - Quite a lot
300
7 - Very Much
……………………………………………………………………………………………………………………
11) How did you find the instructions? Was it easy to follow or do you think it
can be improved?
9) Please provide any suggestions you think could help my study to
improve.
You have done it. Thank you for your participation.
If you want to be entered in the lottery draw to have a chance to win a gift
voucher worth Rs. 500, please enter your name, contact details and email
address below. Two lucky winners will be selected from the lucky draw. Please
be assured these details will solely be used for lottery purpose and will be
destroyed once the draw has been completed.
Name:
Address:
Email address:
301
If you have any comments, complaints or suggestions, use the text box or
contact the address given below.
Contact address:
Moon MoonHalder (PhD Researcher)
Email: [email protected]
Director of studies: Dr. Jens Binder
Email: [email protected]
Nottingham Trent University School of Social Sciences
Burton Street Nottingham
NG1 4BU
302
Debrief Form: Survey
Thank you for your participation! Aims of the Research:
This research looks at the impact of culture on communication behaviour and social identity on Online Social Media.
This study is being conducted in conjunction with an Experiment to investigate how members in a social
media environment can get primed due to environment factors around them which can have an impact
on their cultural and social identity. The data collected will be analysed using statistical methods. We
expect to find that priming will have an impact on culture and communication behaviour. Thank you for
taking part in the study. If you want to know more about this research, have any questions or
suggestions or simply want to find out the progress later please don’t hesitate to contact either me or
my Director of Studies.
Further, if you wish to withdraw your data from this study, please contact me within 14 days of the
study quoting your unique identifier.
If you have queries please contact:
Moon Moon Halder (PhD Researcher) Email: [email protected] Director of studies: Dr. Jens Binder Email: [email protected]
Nottingham Trent University
School of Social Sciences Burton Street Nottingham NG1 4BU UK
303
Appendix 5
Study 4: Individualism version
Used in Indonesia
1. Communication Strategies and Facebook Use in
Indonesia
Thank you for volunteering to participate in this survey.
Before you undertake the survey please read the below information
carefully.
IMPORTANT -- Are you a member of Facebook?
If Yes -- Please read the information carefully and continue with the
survey.
If No -- Sorry you will not be able to participate in the study as this
study also looks at on line activities on Facebook.
304
Participation Information and Consent
Thank you for volunteering to participate in this online survey. Before you undertake the survey please
read the below information carefully.
Aims of the research:
This research looks at your activities and behaviour based on your interactions with your Facebook and non-Facebook (offline) contacts. This study is being undertaken for PhD research.
What will I be asked to do if I agree to take part?
All questions in the survey aim to understand your activities and behaviour based on your interactions with your Facebook and non-Facebook (offline) contacts. After each question you will either be asked to enter your response with a yes/no or a value or choose your best choice from a scale.
The first part is about the number and type of social contacts that you maintain and how you feel about
your social networks. In the second part you will be asked to describe how you are making use of Facebook. In addition, there are some questions about your general behaviour in social contexts. Please read the questions carefully before you respond as once submitted you cannot go back to the previous sections to amend your answers. After you make your choice please rate your response with your best answer. The survey will last approximately 20 - 30 minutes.
Will my data be anonymous?
Yes, your data will be anonymous. Name requirement is optional. However as this is a study on cultural
impact on communication behaviour your ethnicity, gender, age and country of residence, nationality
and first language will be required to make accurate inference from the data collected. Apart from this
no other personal information of identity will be required for this survey.
Summaries of non-personal data (data that can’t be linked to you) such as your responses and
demographic information may be retained long-term as part of a larger data set for publication or
teaching purposes.
Participation in the lottery is your decision for which your name and contact details will need to be
submitted. This information will be kept separate to the survey information and will only be used for
carrying out the lottery.
Do I have to take part?
Your participation in the survey will be entirely your choice. You are free to end the survey at any point
of your participation. If you want that your data be destroyed please contact me or my Director of
Studies within two weeks of completing the survey. My contact information is
[email protected] or my Director of Studies at [email protected] (see full contact
details below).
Important information to consider:
• This study has been ethically approved by the University Research Committee.
• Your responses will be kept anonymous and will be treated as strictly confidential. Responses
collated will be stored in a safe place once the survey is closed.
• The responses will be used for research purposes and the statistical analyses will be used in
conferences and journal articles.
305
If you have queries please contact:
Moon Moon Halder (PhD Researcher) Email: [email protected]
Director of studies: Dr. Jens Binder Email: [email protected] Nottingham Trent University School of Social Sciences Burton Street Nottingham
NG1 4BU
2) Declaration:
By continuing on the next page you agree that the information provided is enough for you to know
about the study and you agree to take part. This is also taken as confirmation that you are 18 or
above.
Please confirm by adding a tick (√) beside each sentence.
5) The study aims and designs has been clearly outlined.
6) I am aware that I can withdraw at any point of the study.
7) I am aware of what I would be expected to do in the study.
8) I am happy to volunteer.
306
2) Demographics:
Please tick (√) or give an answer where applicable:
k) Please add a unique code that will be used for you. This can be a combination of any number
or letter. This is the code that you can use to contact us in case you have any queries or if you
want to withdraw your data.
(e.g., your favourite colour, your favourite letter with any number, etc.)
l) Gender: Please tick (√) the your Gender : Male Female
m) Age: Please provide your Age in years:
n) Ethnicity: Please provide your Ethnicity (e.g., Indian, British, etc.)
o) Work: Please tick the one that is applicable to you.
ii) Employed (full time) ii) Employed (part-time) iii) Unemployed
iv) Student (full time) v) Student (part-time)
p) Country of Residence: Please provide the name of the Country
q) Nationality: What is your Nationality?
r) Town/ City this survey is conducted in: Please provide the name of the Town or City
s) Language: What is your First Language?
t) What was your last English (subject) School/College/ University grade? (e.g., 40, 45, 65.,
etc.)
307
3) In this section please rate the below sentences on your best choice.
Disagree strongly
Disagree
Neither agree or disagree
Agree
Agree strongly
a) I have respect for elders with
whom I interact
b) It is important for me to
maintain harmony within my
social group
c) My happiness depends on the
happiness of those around me
d) I would offer my seat in a bus
to my professor
e) I respect people who are
humble about themselves
f) I will sacrifice my self-interest
for the benefit of the social
group I am in
g) I often have the feeling that my
relationships with others are
more important than my own
accomplishments
h) I should take into consideration
my parents' advice when
making education/ career plans
i) It is important to me to respect
decisions made by my social
group
j) I will stay in a group if they
need me, even when I am not
happy with the group
k) If my brother or sister fails, I
feel responsible
l) Even when I strongly disagree
with my social group members,
I avoid an argument
4) In this section please rate the below sentences on your best choice.
Disagree strongly
Disagree
Neither agree or
disagree
Agree
Agree strongly
a) I'd rather say "No" directly,
than risk being misunderstood
b) Speaking up is not a problem
for me
308
c) Having a lively imagination is
important to me
d) I am comfortable on being
singled out for praise or
rewards
e) I am the same person at home
that I am during social
gathering
f) Being able to take care of
myself is a primary concern to
me
g) I act the same way no matter
who I am with
h) I feel comfortable using
someone's first name soon
after I meet them, even when
they are much older than I am
i) I prefer to be direct and
forthright when dealing with
people I've just met
j) I enjoy being unique and
different from others in many
respects
k) My personal identity is very
important to me
l) I value being in good health
above everything
5) Now please read the traits on the left and rate the 14 traits to the extent to which each trait is
important to you “personally”:
Attributes
Very unimporta
nt
Moderately unimporta
nt
Slightly unimporta
nt
Slightly importa
nt
Moderately important
Very important
a) Respectful b) Independent c) Compliant d) Separate e) Tolerant f) Unconstrained g) Compromising h) Free i) Loyal j) Leader k) Self-sacrificing
l) Unique m) Modest
309
n) Original
6) Which category of people do you best fit in? Please tick your best choice.
a) I value individual achievement and strongly believe in my personal goals.
My personal achievements and success is of priority to me. I strongly believe
in my own efforts. I value independence.
Please select one of the options from below:
1) Not at all
2) A little bit
3) Not sure
4) Quite a lot
5) Very much
b) I value the importance of relationship with others. My goals are achievable/
achieved with the help and support of others around me. Living a life of
harmony with others is of priority to me.
Please select one of the options from below:
1) Not at all
2) A little bit
3) Not sure
4) Quite a lot
5) Very much
310
7) Please read this passage carefully and provide your truthful
response.
Sostoras, a warrior in ancient Sumer, was largely responsible for the success
of their king named Gilgamesh in conquering all of Mesopotamia. As a result,
he was rewarded with a small kingdom of his own to rule.
About 10 years later, Gilgamesh was conscripting warriors for a war. Sostoras
was obligated to send a detachment of soldiers to aid Gilgamesh. He had to
decide who to put in command of the detachment. After thinking about it for
a long time, Sostoras eventually decided on Tiglath who was a talented
general.
This appointment had several advantages. Sostoras was able to make an
excellent general indebted to him. This would solidify Sostoras’s hold on his
own dominion. In addition, the very fact of having a general such as Tiglath
as his personal representative would greatly increase Sostoras's prestige.
Finally, sending his best general would be likely to make Gilgamesh grateful.
Consequently, there was the possibility of getting rewarded by Gilgamesh.
Answer the question:
Do you admire Sostoras? Circle the appropriate answer.
Yes
No
Not sure
311
Pronoun circling task:
8) As a next task concentrate on the text below. Please circle all pronouns
such as I, me, and my and myself that appear in the passage below
and write the total number of pronouns in the box below
I go to the park. My excitement fills me when I see the ice-cream van. I allow
myself to explore every corner of the park, never letting other people distract
me. My voice fills the air and street. I window shop and everywhere I go I see
my reflection looking back at me in the glass from the shops I walk past. When
I return home, my hearts fills with joy and happiness as I know that I will soon
return back in the park. The park belongs to me.
The total number of pronouns in the passage were:
Writing Task:
For the next few minutes please read the questions below and answer the following.
9) List 4 things that you would like to obtain for yourself to improve your
everyday life.
i)…………………….
ii)……………………
iii)…………………..
iv)…………………..
10) List 4 things that you value about yourself as a person.
i)…………………….
312
ii)……………………
iii)…………………..
iv)…………………..
11) Please think of what makes you different from your family and friends.
12) What is your personal goal for the next 1 year?
313
Well done! You have done a great job, just few more to go.
13) There are ten number blanks on the page below. Please write ten answers to the simple
question ‘Who am I?’ in the blanks. Just give twenty different answers to this question.
Answer as if you were giving the answers to yourself, not to somebody else. Write the answers
in order that they occur to you. Don’t worry about the logic or ‘importance’. Go along fairly
fast as time is limited.
21) I am …………………………………………………..
22) I am …………………………………………………..
23) I am …………………………………………………..
24) I am …………………………………………………..
25) I am …………………………………………………..
26) I am …………………………………………………..
27) I am …………………………………………………..
28) I am …………………………………………………..
29) I am …………………………………………………..
30) I am …………………………………………………..
14) Now please read the traits again on the left and rate the 14 traits to the extent to which
each trait is important to you “personally”:
Attributes
Very unimporta
nt
Moderately unimporta
nt
Slightly unimporta
nt
Slightly importa
nt
Moderately important
Very important
a) Respectful b) Independent c) Compliant d) Separate e) Tolerant f) Unconstrained
314
g) Compromising h) Free i) Loyal j) Leader k) Self-sacrificing
l) Unique m) Modest n) Original
15) Please have a think about your Facebook usage and your Facebook social group and rate
the following items/ questions truthfully. Please select only one answer to each question.
Please tick (√) or circle one response per question.
Please read each section and give your best rating:
Please give your best answer Not at all
A little bit
Not sure
Quite a lot
Very much
a. How confident are you that the
information that you upload on Facebook
are not misused by others?
b. How safe do you feel uploading your
pictures and personal information on
Facebook?
c. While on Facebook do you feel you are
at a particular risk?
d. Do you feel that others are at risk
because of using Facebook?
16) Please go through the below traits and rate yourself in comparison to your social
contacts on Facebook.
Social contacts/ social group refers to your social contacts that you have on your Facebook
profile.
I rate myself:
Attributes
Definitely less than my
social contacts on Facebook
Somewhat less than my
social contacts on Facebook
Slightly less than my social
contacts on Facebook
Slightly more
than my social
contacts on
Somewhat more than my social
contacts on Facebook
Definitely more than my social
contacts on Facebook
a) Respectful b) Independent c) Compliant d) Separate
315
e) Tolerant f) Unconstrained g) Compromising h) Free i) Loyal j) Leader k) Self-sacrificing
l) Unique m) Modest n) Original
17. This is in relation to your social contacts on Facebook. Please give your true ratings.
Social contacts/ social group refers to your contacts that you have on your Facebook profile.
Please give your true rating Strongly disagree
Disagree Neither agree or
disagree
Agree Strongly agree
a. I feel confident when my friends appreciate my achievements on Facebook.
b. Sharing harmony among my social
groups on Facebook is crucial to me.
c. Sharing my personal photos and information on Facebook gives me a sense of freedom.
d. I feel valued and appreciated when my
friends share their likes and comments on my personal photos and information on Facebook.
f. My popularity on Facebook depends on the number of friends I have on Facebook.
18. Now think about the social support that you receive and give to your social group
on Facebook. Please read the sentences below carefully and give your truthful ratings.
Social contacts/ social group refers to your contacts that you have on your Facebook profile.
Please give your true rating Extremely unlikely
Unlikely Neutral Likely Extremely likely
a. How likely would you offer social
support to others?
b. How likely would you share your most private worries and fears with someone in your social group on Facebook?
316
c. How likely do you feel that members
of your social group would help you during a personal crisis?
d. How likely are you to involve yourself with a social cause on Facebook?
19. The below sentences talks about you and your relationships that you share with
your social group on Facebook. Please read the below sentences and give your truthful
ratings.
Social contacts/ social group refers to your contacts that you have on your Facebook profile.
Please give your true rating
Never Rarely Sometimes Often Always
a. I feel strongly connected with my social
group members on Facebook
b. I often feel held back by my social group on Facebook
c. My Facebook friends are very important
for me
d. I sometimes make excuses of belonging to my friends group on Facebook
20. Now think about your Facebook contacts and how much you trust them. Read the
sentences below and give your truthful ratings.
Social contacts/ social group refers to your contacts that you have on your Facebook profile.
Please give your true rating
Never Not so often
Sometimes Often Very often
a. My Facebook social group members are
trustworthy
b. I am confident that my friends on Facebook mostly share their honest opinions
about me
c. I trust Facebook for safeguarding my
personal information
d. Members of my social group on Facebook
does their best to help me whenever I need them
317
Scenarios:
21. How likely are you to act in a particular way?
Below are some scenarios that you often come across while using Facebook. Please read the
scenarios carefully and give your truthful answers.
Social contacts/ social group refers to your contacts that you have on your Facebook profile.
How likely are you to act in a particular way?
Please give your truthful answers:
Extremely unlikely
Unlikely Neutral Likely Extremely likely
a. I log on my Facebook and I receive a
friend request from my family member. How likely is that you will accept the friend request?
b. I log on my Facebook and receive a friend request from an unknown person. How likely will you will accept the friend request?
c. While logging on Facebook I am reminded to update my privacy settings.
How likely would you update your privacy settings.
d. While on Facebook I get a request to join a protest for a noble/ social cause. How likely would you join the protest.
e. While on Facebook when I see my
friends talk about their achievements. I would congratulate them.
f. While on Facebook I often seem to join social groups that has personal
importance to me.
22. Now think back on the whole survey that you have just completed and give your ratings
based on the overall understanding of the surveys questions. Please give your truthful
answers as this will help me to spot areas of improvement in my study:
What do you think this study put a focus on?
What do you think this study
put a focus on?
Not at
all
A
little bit
Somewhat Not
sure/ N/A
Quite
a bit
Quite
a lot
Very
much
a. Social Responsibility b. Individual decisions c. Reading skills
318
23. Did you find the instructions easy to follow?
Please put a circle or add a tick (√) to your choice. If you select “No” please give a reason.
i) Yes ii) Not sure iii) No
24. Do you think there is room for development with the study design or the
questions in general?
Please put a circle or adding a tick (√) to your choice. If you select “Yes” or “Not sure” please
give a reason.
i) Yes ii) Not sure iii) No
319
You have done it. Thank you for your participation.
If you have any comments, complaints or suggestions, use the text box or contact the
address given below. Please provide your unique id code (optional)
320
Debrief Form: Survey
Thank you for your participation! Aims of the Research:
This research looks at the impact of culture on communication behaviour and social identity on Online Social Media.
This study is being conducted in conjunction with an Experiment to investigate how members in a social
media environment can get primed due to environment factors around them which can have an impact
on their cultural and social identity. The data collected will be analysed using statistical methods. We
expect to find that priming will have an impact on culture and communication behaviour. Thank you for
taking part in the study. If you want to know more about this research, have any questions or
suggestions or simply want to find out the progress later please don’t hesitate to contact either me or
my Director of Studies.
Further, if you wish to withdraw your data from this study, please contact me within 14 days of the
study quoting your unique identifier.
If you have queries please contact:
Moon Moon Halder (PhD Researcher) Email: [email protected] Director of studies: Dr. Jens Binder Email: [email protected]
Nottingham Trent University
School of Social Sciences Burton Street Nottingham NG1 4BU UK
321
Appendix 6
Study 4: Collectivism version
Used in Indonesia
2. Communication Strategies and Facebook Use in
Indonesia
Thank you for volunteering to participate in this survey.
Before you undertake the survey please read the below information
carefully.
IMPORTANT -- Are you a member of Facebook?
If Yes -- Please read the information carefully and continue with the
survey.
If No -- Sorry you will not be able to participate in the study as this
study also looks at on line activities on Facebook.
322
Participation Information and Consent
Thank you for volunteering to participate in this online survey. Before you undertake the survey please
read the below information carefully.
Aims of the research:
This research looks at your activities and behaviour based on your interactions with your Facebook and non-Facebook (offline) contacts. This study is being undertaken for PhD research.
What will I be asked to do if I agree to take part?
All questions in the survey aim to understand your activities and behaviour based on your interactions with your Facebook and non-Facebook (offline) contacts. After each question you will either be asked to enter your response with a yes/no or a value or choose your best choice from a scale.
The first part is about the number and type of social contacts that you maintain and how you feel about
your social networks. In the second part you will be asked to describe how you are making use of Facebook. In addition, there are some questions about your general behaviour in social contexts. Please read the questions carefully before you respond as once submitted you cannot go back to the previous sections to amend your answers. After you make your choice please rate your response with your best answer. The survey will last approximately 20 - 30 minutes.
Will my data be anonymous?
Yes, your data will be anonymous. Name requirement is optional. However as this is a study on cultural
impact on communication behaviour your ethnicity, gender, age and country of residence, nationality
and first language will be required to make accurate inference from the data collected. Apart from this
no other personal information of identity will be required for this survey.
Summaries of non-personal data (data that can’t be linked to you) such as your responses and
demographic information may be retained long-term as part of a larger data set for publication or
teaching purposes.
Participation in the lottery is your decision for which your name and contact details will need to be
submitted. This information will be kept separate to the survey information and will only be used for
carrying out the lottery.
Do I have to take part?
Your participation in the survey will be entirely your choice. You are free to end the survey at any point
of your participation. If you want that your data be destroyed please contact me or my Director of
Studies within two weeks of completing the survey. My contact information is
[email protected] or my Director of Studies at [email protected] (see full contact
details below).
Important information to consider:
• This study has been ethically approved by the University Research Committee.
• Your responses will be kept anonymous and will be treated as strictly confidential. Responses
collated will be stored in a safe place once the survey is closed.
• The responses will be used for research purposes and the statistical analyses will be used in
conferences and journal articles.
323
If you have queries please contact:
Moon Moon Halder (PhD Researcher) Email: [email protected]
Director of studies: Dr. Jens Binder Email: [email protected] Nottingham Trent University School of Social Sciences Burton Street Nottingham
NG1 4BU
3) Declaration:
By continuing on the next page you agree that the information provided is enough for you to know
about the study and you agree to take part. This is also taken as confirmation that you are 18 or
above.
Please confirm by adding a tick (√) beside each sentence.
9) The study aims and designs has been clearly outlined.
10) I am aware that I can withdraw at any point of the study.
11) I am aware of what I would be expected to do in the study.
12) I am happy to volunt
2) Demographics:
Please tick (√) or give an answer where applicable:
u) Please add a unique code that will be used for you. This can be a combination of any number
or letter. This is the code that you can use to contact us in case you have any queries or if you
want to withdraw your data.
(e.g., your favourite colour, your favourite letter with any number, etc.)
v) Gender: Please tick (√) the your Gender : Male Female
324
w) Age: Please provide your Age in years:
x) Ethnicity: Please provide your Ethnicity (e.g., Indian, British, etc.)
y) Work: Please tick the one that is applicable to you.
iii) Employed (full time) ii) Employed (part-time) iii) Unemployed
iv) Student (full time) v) Student (part-time)
z) Country of Residence: Please provide the name of the Country
aa) Nationality: What is your Nationality?
bb) Town/ City this survey is conducted in: Please provide the name of the Town or City
cc) Language: What is your First Language?
dd) What was your last English (subject) School/College/ University grade? (e.g., 40, 45, 65.,
etc.)
325
3) In this section please rate the below sentences on your best choice.
Disagree strongly
Disagree
Neither agree or disagree
Agree
Agree strongly
a) I have respect for elders with
whom I interact
b) It is important for me to
maintain harmony within my
social group
c) My happiness depends on the
happiness of those around me
d) I would offer my seat in a bus
to my professor
e) I respect people who are
humble about themselves
f) I will sacrifice my self-interest
for the benefit of the social
group I am in
g) I often have the feeling that my
relationships with others are
more important than my own
accomplishments
h) I should take into consideration
my parents' advice when
making education/ career plans
i) It is important to me to respect
decisions made by my social
group
j) I will stay in a group if they
need me, even when I am not
happy with the group
k) If my brother or sister fails, I
feel responsible
l) Even when I strongly disagree
with my social group members,
I avoid an argument
4) In this section please rate the below sentences on your best choice.
Disagree strongly
Disagree
Neither agree or disagree
Agree
Agree strongly
a) I'd rather say "No" directly,
than risk being misunderstood
326
b) Speaking up is not a problem
for me
c) Having a lively imagination is
important to me
d) I am comfortable on being
singled out for praise or
rewards
e) I am the same person at home
that I am during social
gathering
f) Being able to take care of
myself is a primary concern to
me
g) I act the same way no matter
who I am with
h) I feel comfortable using
someone's first name soon
after I meet them, even when
they are much older than I am
i) I prefer to be direct and
forthright when dealing with
people I've just met
j) I enjoy being unique and
different from others in many
respects
k) My personal identity is very
important to me
l) I value being in good health
above everything
5) Now please read the traits on the left and rate the 14 traits to the extent to which each trait is
important to you “personally”:
Attributes
Very unimporta
nt
Moderately unimporta
nt
Slightly unimporta
nt
Slightly importa
nt
Moderately important
Very important
a) Respectful b) Independent c) Compliant d) Separate e) Tolerant f) Unconstrained g) Compromising h) Free i) Loyal j) Leader k) Self-sacrificing
327
l) Unique m) Modest n) Original
6) Which category of people do you best fit in? Please tick your best choice.
a) I value individual achievement and strongly believe in my personal goals.
My personal achievements and success is of priority to me. I strongly believe
in my own efforts. I value independence.
Please select one of the options from below:
1) Not at all
2) A little bit
3) Not sure
4) Quite a lot
5) Very much
b) I value the importance of relationship with others. My goals are achievable/
achieved with the help and support of others around me. Living a life of
harmony with others is of priority to me.
Please select one of the options from below:
1) Not at all
2) A little bit
3) Not sure
4) Quite a lot
5) Very much
328
7) Please read this passage carefully and provide your truthful
response.
Sostoras, a warrior in ancient Sumer, was largely responsible for the success
of their king named Gilgamesh in conquering all of Mesopotamia. As a result,
he was rewarded with a small kingdom of his own to rule.
About 10 years later, Gilgamesh was conscripting warriors for a war. Sostoras
was obligated to send a detachment of soldiers to aid Gilgamesh. He had to
decide who to put in command of the detachment. After thinking about it for
a long time, Sostoras eventually decided on Tiglath who was a member of his
family.
This appointment had several advantages. Sostoras was able to show his
loyalty towards his family. He was also able to cement his loyalty to him. In
addition, having Tiglath as a commander increased the power and prestige of
his family.
Answer the question:
Do you admire Sostoras? Circle the appropriate answer.
Yes
No
Not sure
329
Pronoun circling task:
8) As a next task concentrate on the text below. Please circle all pronouns
such as I, me, and my and myself that appear in the passage below
and write the total number of pronouns in the box below
We go to the park. Our excitement fills us when we see the ice-cream van.
We allow ourselves to explore every corner of the park, never letting other
people distract us. Our voice fills the air and street. We window shop and
everywhere we go we see our reflections looking back at us in the glass from
the shops we walk past. When we return home, our hearts fills with joy and
happiness as we know that we will soon return back in the park. The park
belongs to us.
The total number of pronouns in the passage were:
Writing Task:
For the next few minutes please read the questions below and answer the following.
9) List 4 things that you would like to obtain for yourself to improve your
everyday life.
i)…………………….
ii)……………………
iii)…………………..
iv)…………………..
10) List 4 things that you value about yourself as a person.
330
i)…………………….
ii)……………………
iii)…………………..
iv)…………………..
11) Please think of what makes you different from your family and friends.
12) What is your personal goal for the next 1 year?
331
Well done! You have done a great job, just few more to go.
13) There are ten number blanks on the page below. Please write ten answers to the simple
question ‘Who am I?’ in the blanks. Just give twenty different answers to this question.
Answer as if you were giving the answers to yourself, not to somebody else. Write the answers
in order that they occur to you. Don’t worry about the logic or ‘importance’. Go along fairly
fast as time is limited.
31) I am …………………………………………………..
32) I am …………………………………………………..
33) I am …………………………………………………..
34) I am …………………………………………………..
35) I am …………………………………………………..
36) I am …………………………………………………..
37) I am …………………………………………………..
38) I am …………………………………………………..
39) I am …………………………………………………..
40) I am …………………………………………………..
14) Now please read the traits again on the left and rate the 14 traits to the extent to which
each trait is important to you “personally”:
Attributes
Very unimporta
nt
Moderately unimporta
nt
Slightly unimporta
nt
Slightly importa
nt
Moderately important
Very important
a) Respectful b) Independent c) Compliant d) Separate e) Tolerant f) Unconstrained
332
g) Compromising h) Free i) Loyal j) Leader k) Self-sacrificing
l) Unique m) Modest n) Original
15) Please have a think about your Facebook usage and your Facebook social group and rate
the following items/ questions truthfully. Please select only one answer to each question.
Please tick (√) or circle one response per question.
Please read each section and give your best rating:
Please give your best answer Not at all
A little bit
Not sure
Quite a lot
Very much
a. How confident are you that the
information that you upload on Facebook
are not misused by others?
b. How safe do you feel uploading your
pictures and personal information on
Facebook?
c. While on Facebook do you feel you are
at a particular risk?
d. Do you feel that others are at risk
because of using Facebook?
16) Please go through the below traits and rate yourself in comparison to your social
contacts on Facebook.
Social contacts/ social group refers to your social contacts that you have on your Facebook
profile.
I rate myself:
Attributes
Definitely less than my
social contacts on Facebook
Somewhat less than my
social contacts on Facebook
Slightly less than my social
contacts on Facebook
Slightly more
than my social
contacts on
Somewhat more than my social
contacts on Facebook
Definitely more than my social
contacts on Facebook
a) Respectful b) Independent c) Compliant d) Separate
333
e) Tolerant f) Unconstrained g) Compromising h) Free i) Loyal j) Leader k) Self-sacrificing
l) Unique m) Modest n) Original
17. This is in relation to your social contacts on Facebook. Please give your true ratings.
Social contacts/ social group refers to your contacts that you have on your Facebook profile.
Please give your true rating Strongly disagree
Disagree Neither agree or
disagree
Agree Strongly agree
a. I feel confident when my friends appreciate my achievements on Facebook.
b. Sharing harmony among my social
groups on Facebook is crucial to me.
c. Sharing my personal photos and information on Facebook gives me a sense of freedom.
d. I feel valued and appreciated when my
friends share their likes and comments on my personal photos and information on Facebook.
f. My popularity on Facebook depends on the number of friends I have on Facebook.
18. Now think about the social support that you receive and give to your social group
on Facebook. Please read the sentences below carefully and give your truthful ratings.
Social contacts/ social group refers to your contacts that you have on your Facebook profile.
Please give your true rating Extremely unlikely
Unlikely Neutral Likely Extremely likely
a. How likely would you offer social
support to others?
b. How likely would you share your most private worries and fears with someone in your social group on Facebook?
334
c. How likely do you feel that members
of your social group would help you during a personal crisis?
d. How likely are you to involve yourself with a social cause on Facebook?
19. The below sentences talks about you and your relationships that you share with
your social group on Facebook. Please read the below sentences and give your truthful
ratings.
Social contacts/ social group refers to your contacts that you have on your Facebook profile.
Please give your true rating
Never Rarely Sometimes Often Always
a. I feel strongly connected with my social
group members on Facebook
b. I often feel held back by my social group on Facebook
c. My Facebook friends are very important
for me
d. I sometimes make excuses of belonging to my friends group on Facebook
20. Now think about your Facebook contacts and how much you trust them. Read the
sentences below and give your truthful ratings.
Social contacts/ social group refers to your contacts that you have on your Facebook profile.
Please give your true rating
Never Not so often
Sometimes Often Very often
a. My Facebook social group members are
trustworthy
b. I am confident that my friends on Facebook mostly share their honest opinions
about me
c. I trust Facebook for safeguarding my
personal information
d. Members of my social group on Facebook
does their best to help me whenever I need them
335
Scenarios:
21. How likely are you to act in a particular way?
Below are some scenarios that you often come across while using Facebook. Please read the
scenarios carefully and give your truthful answers.
Social contacts/ social group refers to your contacts that you have on your Facebook profile.
How likely are you to act in a particular way?
Please give your truthful answers:
Extremely unlikely
Unlikely Neutral Likely Extremely likely
a. I log on my Facebook and I receive a
friend request from my family member. How likely is that you will accept the friend request?
b. I log on my Facebook and receive a friend request from an unknown person. How likely will you will accept the friend request?
c. While logging on Facebook I am reminded to update my privacy settings.
How likely would you update your privacy settings.
d. While on Facebook I get a request to join a protest for a noble/ social cause. How likely would you join the protest.
e. While on Facebook when I see my
friends talk about their achievements. I would congratulate them.
f. While on Facebook I often seem to join social groups that has personal
importance to me.
22. Now think back on the whole survey that you have just completed and give your ratings
based on the overall understanding of the surveys questions. Please give your truthful
answers as this will help me to spot areas of improvement in my study:
What do you think this study put a focus on?
What do you think this study
put a focus on?
Not at
all
A
little bit
Somewhat Not
sure/ N/A
Quite
a bit
Quite
a lot
Very
much
a. Social Responsibility b. Individual decisions c. Reading skills
336
23. Did you find the instructions easy to follow?
Please put a circle or add a tick (√) to your choice. If you select “No” please give a reason.
i) Yes ii) Not sure iii) No
24. Do you think there is room for development with the study design or the
questions in general?
Please put a circle or adding a tick (√) to your choice. If you select “Yes” or “Not sure” please
give a reason.
i) Yes ii) Not sure iii) No
You have done it. Thank you for your participation.
If you have any comments, complaints or suggestions, use the text box or contact the
address given below. Please provide your unique id code (optional)
337
Debrief Form: Survey
Thank you for your participation! Aims of the Research:
This research looks at the impact of culture on communication behaviour and social identity on Online Social Media.
This study is being conducted in conjunction with an Experiment to investigate how members in a social
media environment can get primed due to environment factors around them which can have an impact
on their cultural and social identity. The data collected will be analysed using statistical methods. We
expect to find that priming will have an impact on culture and communication behaviour. Thank you for
taking part in the study. If you want to know more about this research, have any questions or
suggestions or simply want to find out the progress later please don’t hesitate to contact either me or
my Director of Studies.
Further, if you wish to withdraw your data from this study, please contact me within 14 days of the
study quoting your unique identifier.
If you have queries please contact:
Moon Moon Halder (PhD Researcher) Email: [email protected] Director of studies: Dr. Jens Binder Email: [email protected]
Nottingham Trent University
School of Social Sciences Burton Street Nottingham NG1 4BU UK
338
Appendix 7
Figure 5.5
Model 9 & 10 Illustrates Standardized Path Coefficients for the four Mediators using the
combined models for Online Group Identity for each cultural orientation
Note: FB = Facebook; FB Active = active interaction on Facebook; FB Passive = passive interaction on Facebook; FB Days = average number of
days’ respondents had been on Facebook in the last 14 days and FB HR = average number of time (in minutes) respondents had been on Facebook
in the last 14 days
339
Appendix 8
Figure 5.10
Models 17 & 18 Illustrates Standardized Path Coefficients for the four Mediators using the
combined models for each cultural orientation on the Collectivist attributes for Online Self-
Enhancement
Note: FB = Facebook; FB Active = active interaction on Facebook; FB Passive = passive interaction on Facebook; FB Days = average number of
days’ respondents had been on Facebook in the last 14 days and FB HR = average number of time (in minutes) respondents had been on Facebook
in the last 14 days
340
Appendix 9
Figure 5.11
Models 19 & 20 Illustrates Standardized Path Coefficients for the four Mediators using the
combined models for each cultural orientation on the Individualist attributes for Online Self-
Enhancement
Note: FB = Facebook; FB Active = active interaction on Facebook; FB Passive = passive interaction on Facebook; FB Days = average number of
days’ respondents had been on Facebook in the last 14 days and FB HR = average number of time (in minutes) respondents had been on Facebook
in the last 14 days
341
Appendix 10
Figure 5.14
Models 9 & 10 Illustrates Standardized Path Coefficients for the four Mediators using the
combined models for each cultural orientation on Perceived Online Social Support
FB = Facebook; FB Active = active interaction on Facebook; FB Passive = passive interaction on Facebook; FB Days = average number of days’
respondents had been on Facebook in the last 14 days and FB HR = average number of time (in minutes) respondents had been on Facebook in
the last 14 days
342
Appendix 11
Figure 5.17
Model 9 & 10 Illustrates Standardized Path Coefficients for the four Mediators using the
combined models for Online Trust for each cultural orientation
FB = Facebook; FB Active = active interaction on Facebook; FB Passive = passive interaction on Facebook; FB Days = average number of days’
respondents had been on Facebook in the last 14 days and FB HR = average number of time (in minutes) respondents had been on Facebook in
the last 14 days
343
Appendix 12
Figure 5.19
Model 9 & 10 Illustrates Standardized Path Coefficients for the four Mediators using the
combined models for Online Privacy Concern for each cultural orientation
FB = Facebook; FB Active = active interaction on Facebook; FB Passive = passive interaction on Facebook; FB Days = average number of days’
respondents had been on Facebook in the last 14 days and FB HR = average number of time (in minutes) respondents had been on Facebook in
the last 14 days
344
Appendix 13
13.1. Challenges faced during the Cross-cultural Data Collection
This section is part of a published journal article (Halder et al., 2016).
Cross-cultural research helps to study behaviour in different cultures which helps researchers
to make valid inferences of their findings. However, the experiences encountered conducting
cross-cultural research might not be the same for all. It is believed that by highlighting the
experiences encountered during this cross-cultural research would help future researchers could
benefit from the recommendations provided as they can adequately prepare themselves in advance.
This section is part of a published journal article (Halder et al., 2016). The main challenges
experienced were in the areas of survey design, translation, data collection, cultural obligation and
peer pressure, ethical consideration and awareness, experiences of working with a cross-cultural
team and the issues faced as a researcher at a personal level will be discussed below, followed by
a discussion on the wider effects of cross-cultural research, its future implications and
recommendations and finally a conclusion of this section.
13.2. Survey Design
The study was in the form of a survey adapted from previous studies into culture-specific
attitudes and behaviours in different social situations (Gardner, Gabriel, & Lee, 1999; Oyserman
& Lee, 2008). The scales used in the survey were validated scales, however they had been
developed keeping in mind Western theories and had been mostly used on Western samples.
Therefore, it was important that item equivalence was maintained in both cultures in order to
identify any “true” cultural differences. Such equivalence in instrument design can only be
achieved when researchers are mindful of the various idioms, phrases and grammatical detail to
be found in a particular location and, more generally, how respondents make inferences in different
cultures (Sekaran, 1983). For example, while “Feeling guilty for my brother’s/sister’s failure” was
one of the item in the survey. Such feelings might be “strongly relevant” to respondents in
Collectivists countries where members believe in being part of their social structure and as such
might hold themselves responsible for not being able to guide or support their brother/sister which
345
could have prevented their failure. However, such feelings might be less agreeable in Individualist
countries where members believe in being responsible for their own actions and behaviors. The
inferences individuals make are highly influenced by cultural background one come from which
influences how we think, perceive and react to situations around us (Cunningham et al., 1995;
Dake, 1991; Kühnen & Oyserman, 2002; Oyserman & Lee, 2008). Therefore, it is important to
point out that participant responses are influenced by their cultural background they come from
and in order to interpret the results correctly, researchers need to have a culturally fit research
instrument and they should also possess an understanding of their culture.
13.3. Translation
Due to the cultural variability in the study, it was not possible to use a single instrument that
could be consistently used in both cultures (Sechrest et al., 1972; Sekaran, 1983). Therefore, it was
decided to administer the survey in the native langue of the participants in India and Indonesia
which was achieved by back-translation of the survey items. It was however made sure that both
the surveys had the same literal meaning (Brislin, 1970; McGorry, 2000). Choosing the most
suitable translation method was also important so that it limited any chances of translation errors.
Although, there were possibilities of using more than one translation method to achieve a more
accurate and culturally fit instrument (McGorry, 2000) however this was not always possible due
to restrictions in time and funding. Care was also taken that the participants in both the cultures
received the same information (Sekaran, 1983).
13.4. Data Collection
The level of familiarity with the general research process and participation in research studies
among the Indian sample was certainly a concern as not all universities in India indulged in similar
research activities. This called for developing creative ways to administer the surveys in both
cultures so that all participants could fully understand the participation process and their rights.
All the students were from Higher Education and had good level of English both written and verbal
although it was difficult to find out what their actual level of understanding was. It was observed
346
that in spite of their familiarity with English language, they still had difficulty in following the
overall study participation process. Here, this issue was resolved as the researcher was present in
person along with one of the bilingual translators to assist participants while taking the survey.
However, in other cases such as online surveys participants’ understanding of the survey questions
might be restricted and participants might end up responding incorrectly to the questions, i.e., in
ways they would not follow with a better understanding of the study. Unintended responses will
certainly have an adverse impact on the research outcome and may produce an effect when actually
there is none. Global demand and use of the internet has made researchers change and adapt to
newer ways of conducting their research. In particular, for cross-cultural research online methods
are attractive as they save time and are also cost effective. However, the absence of personal cues
and support might also have a negative impact on the quality of the data collected.
Conducting cross-cultural research also required planning ahead. For example, the time of year
when the research would be conducted in a country. While particular dates and times might be
useful and convenient for the researcher in one country, it might not be the same in another country.
The data collection process in the UK was conducted without any hindrances, whereas a different
picture emerged in India as national holidays were suddenly called for by the government due to
the election as there were social unrest in some parts of the country. As a result, there were only
limited numbers of students at the institution during the data collection phase which had a negative
impact on the sample size. Therefore, it is suggested that although planning ahead is always useful,
researchers should always plan for sudden changes. Planning ahead can include aspects like
allocating additional meeting times, checking university opening and closing times and also
identifying the most promising time for data collection.
Informal meetings with Indian students after their participation in the survey suggested that
although they were anxious about the social situation in the city, they felt that as students they
were obligated to participate in the survey as it has been requested by their lecturers. Such loyalty
and compromising behaviours are a part of collectivist cultures (Heine & Dehman, 1997; Markus
& Kitayama, 1991) and it can be suggested that being loyal to their social group helped participants
to self-enhance (Gaertner, Sedikides, & Chang, 2008) rather having a negative experience.
However, this raises the question of ethical implications as discussed further down.
347
13.5. Cultural obligation & Peer pressure
Having analysed the data and looked at the statistical results, further questions had to be
addressed. While the survey results followed the expected trend in the UK population, results
obtained from the Indian sample had some out of norm results and were not in line with what
would be expected in that cultural context. The difference in results could be a product of the
impact of globalization which has resulted in the change in attitudes and behaviours of Indian
respondents. On the other hand, it could also be assumed that the scales used in the survey, which
as mentioned earlier had been originally developed keeping in mind western attitudes and
behaviours (Singelis, 1994; Trafimow, Triandis, & Goto, 1991) was not really applicable for
collectivist culture like India. It is also important to address that participants (all students at Higher
Education institutions) in India showed an obligatory role towards their seniors, such as their
lecturers and Deans in the institution, which could also be one of the contributing factors to the
results obtained. Participants may have simply responded without actually fully comprehending
the meaning of the questions.
Deans and lecturers hold high positions in the social hierarchical system in collectivist countries
where teaching roles carry particular authority. Maintaining harmony and loyalty towards others,
especially individuals who hold higher social positions, is regarded as an obligatory factor in a
collectivist culture (Basabe & Ros, 2005; Markus & Kitayama, 1991). While such collectivist
behavior may have helped participants to self-enhance and to improve their self-esteem and
subjective well-being (Kurman, 2003; Sedikides, Gaertner, & Toguchi, 2003), it can also be argued
that because of such social pressure some of the Indian participants might have taken part in the
survey against their wishes. This can have adverse ramification for the survey results as
participants might not have given their truthful responses. Additionally, this also fails to maintain
ethical standards.
Participants in the Indian institutions were also under constant peer pressure. Informal
discussions after the survey indicated that some of the participants took part in the survey to
maintain group harmony as it contributed to collective action whether they liked it or not. They
believed in following the actions of their group members as they feel obligated to be loyal to their
in-group. A group context, e.g., a class room, in combination with peer pressure can also make
participants more biased in their responses as they might respond in accordance with group norms
348
and not their individual opinion. While the participants in the UK were direct and independent in
their participation, Indian participants were more indirect in their approach. For example, Indian
participants raised questions only when approached by the researcher unlike UK participants who
asked questions whenever they wanted to clarify anything during the study. While such differences
in behaviour are simply a reflection of cultural backgrounds (Triandis, 1989), they do highlight
the importance of the presence of personal cues during data collection.
13.6. Ethical Consideration and Awareness
All research involving human participation calls for maintaining ethical standards and
following ethical guidelines. This is even more important in cross-cultural research as identifying
and understanding the cultural specific variance in a sample is of utmost importance. Issues like
anonymity, participant information and informed consent, information about the data collection
process and information of ownership of the data are some of the points that should always be
considered. These points also provide a professional context in which participants give information
to researchers. Ethical codes and practices that is followed do not always address all ethical issues
that researchers might encounter. However, by being aware of the values, norms, perception and
behaviours in the target culture such issues can be addressed (Ponterotto & Casas, 1993). It is also
important to understand that ethical codes and practices that are developed within western
countries cannot fully be followed when conducting research in Eastern countries (Pedersen,
1991). This follows from some fundamental and well-documented differences in cultural norms,
values, and behavioural patterns (Hofstede, 1980; Hofstede, Hofstede, & Minkov, 1997).
Therefore, having an understanding of how ethical procedures may be received in the cultures
under investigation is vital in administering the study and also in the rightful interpretation of the
data.
For the present research, ethics and cultural differences had to be considered from early on in
the research process. The UK sample included students at UK institutions, who had full access to
the internet at all times and were fully aware of issues surrounding data collection and ethical
standards. The survey was therefore circulated both online and in paper format to maximize the
number of participants. The online survey was distributed through the institution’s research
349
participation site which also made it convenient for both the participants and the researcher.
However, not all students at the Indian institutions would have access to the Internet and have a
thorough understanding of the ethical guidelines for research. Hence the survey in India was
conducted in paper format in both English and Bengali. All participants were made aware, as far
as possible, of the ethical guidelines and participant information was provided which included their
rights as a research participant before the survey was circulated. Therefore, during the conduct of
any cross-cultural research awareness of cultural history, values and traditions are essential which
needs to be incorporated in the ethical guidelines for the researchers.
13.7. Experience of working as a Cross-Cultural Team
Having the support, flexibility of the teams in UK and in India were really beneficial as they
helped by providing their guidance under any doubtful situation. They not only helped to adapt the
research strategy according to various changing situations but also provided guidance throughout
the process. While the team in UK was focused in getting the data collection completed on time,
the team in India displayed quite a laid back approach. This could have been due to actual cultural
differences, or it could have been due to a lack of personal interest given that research outcomes
were directly relevant to researchers in the UK, not India.
There was a clearly discernible lack of knowledge of research ethics among Indian students.
While the lecturers did possess knowledge about ethical practices that they were required to follow,
such information was clearly not passed on to their students. While there was a consensus in both
the teams (UK and India) on data protection, there was certainly a difference in their approach to
this issue. For example, while anonymization of data and safe storage of data were a given
requirement in the UK, the team in India did not seem to be overly concerned about data handling
or storage. What was clear from the Indian approach was a need to help and support in the data
collection process which certainly were collectivist behaviours considering that maintaining a
harmonious relationship is one of the primary goals of the members of collectivist cultures, even
if it means sacrificing personal interests (Triandis, 1989). Furthermore, as this research was
referred by a personal contact to the Indian team, it is quite likely that offering to help in the data
350
collection process meant fulfilling a favour and maintaining a harmonious relationship with
personal contacts.
13.8. Issues faced as a Researcher at a Personal level
At the initial stage of the research in India there were substantial delays due to failed response
from the contacts that was initially established from the UK. Although email exchanges and
telephone conversations had taken place before planning the visit to India where access to student
participants at several universities was promised, this was not fulfilled after the study started in
India. This sudden and unexpected delay had a demotivating psychological impact. However, with
the help of other established contacts in India, which had to be planned suddenly helped to liaise
with universities and eventually get access to research participants. This sudden delay in my data
collection in a different country could have had knock-on effect on my research on the whole. It
is therefore advisable that while conducting cross-cultural research, it is important to engage in
ongoing communication with not just one but several contacts so that the researcher is not
dependent on anyone in particular as there is no guarantee that a particular social contact will work.
It is also advisable to have enough time at hand to adapt to unexpected changes in the research
plan as otherwise any adverse impact on the overall research cannot be cushioned.
One of the ways to help adapt to changes is to have a positive attitude. For example, in spite of
the sudden delay in the data collection and also the social support that received from family, friends
and my team in the UK helped motivated to maintain focus in the data collection. While it is not
always possible to have the privilege of family and friends during a cross-cultural data collection,
it is always helpful to get to know and establish sound relationships with locals in the area as they
can be helpful during any unexpected circumstances.
Keeping a reflexive diary of daily activities during the research would also help to formulate
plans in advance should you need to. The daily reflexive notes during the data collection had
helped not only to keep track of the progress and meet deadlines, but also to figure out daily
strategies in advance. For example, on more than one occasion some of the universities failed to
get provide access to research participants due to which other contacts were approached during the
visit. The reflexive notes helped to decide whether waiting for the response from a particular
351
university was worthwhile or additional contacts needed to be approached in order to complete the
data collection on time. Making use of new opportunities that come up can always be a good idea
as it reduces the risk of being dependable on just one contact.
13.9. Wider Effects of Cross-Cultural Research
This particular study included an actual experiment. The survey came in two different versions
with the aim of making participants switch between cultural identities. This switch was expected
to last at least for the time it took participants to respond to the remainder of the survey. As such,
some manipulation of participants was taking place. The survey was considered safe to be used in
both cultures as it did not aim to reverse participant behaviour but to highlight different aspects of
their existing self-concept to them. No adverse psychological impact on the participants could be
expected from the survey. But on a more general level this begs the question whether all research
methods are equally appropriate for use in different cultures.
The selection of method in cross-cultural research is very crucial as implications of incorrect
methodological procedures can sometimes lead to adverse psychological and social issues for the
participants in certain countries. While one method of study might be easy to use and implement
in one country, it might not be the same in another country. For example, while a survey response
of participants on views on adult images in newspapers might be easy to accept in an individualist
country, such topics might not be easily accepted in collectivist countries. Therefore, it is important
for the researcher to anticipate the culture-specific impact of the method itself. One solution could
be the consultation of country-specific ethical boards before data collection. Obviously this
depends on the availability of such boards. A researcher firmly embedded in one particular culture
may not always be able to assess the psychological impact of a piece of research on participants in
another culture. As researchers we can only be cautious and take measures to eliminate any
possibilities of adverse psychological impact on our participants.
352
13.10. Future Implications and Recommendations
Recognition and adaptability to different cultural norms, values and behaviours is called for at
numerous points in the research process. A well planned research strategy and flexible
methodological approach should be incorporated in any cross-cultural research. Research
instruments require to be appropriately back translated in the native language in order to be viable
both culturally and literally which will not only help the participants to be able to comprehend the
meaning of the items but will also enable them to respond truthfully and correctly. Failing to get
the meaning of the items correct will have an adverse impact on the data collected and as such
researchers should be cautious when interpreting such data.
Cultural understanding and familiarity with cultural specific norms and behaviours would help
researchers develop contacts more easily and conduct studies in a time-efficient manner. It is also
recommended that ethical guidelines should be seen from a cross-cultural perspective and should
take into consideration cultural differences while formulating or implementing ethical guidelines
and practices. Some ethical guidelines and practices which have been developed for research in
western countries may be difficult to incorporate and implement in eastern countries. This calls
for developing culturally appropriate guidelines and practices. It is recommended that researchers
conduct a pilot study prior to any main study to test their research strategy and their instruments.
It is also recommended that participant feedback is collected after the study. This can be a part of
the study at the end or researchers can also have informal discussions with the participants in
person which will help to identify areas of improvement or amendment.
13.11. Summary: Cross-Cultural Challenges
This section was an attempt to discuss the experiences that had been experienced as a researcher
conducting a cross-cultural research. It provided an overview of some of the challenges that
researchers can face while conducting cross-cultural research and recommended ways how such
challenges could be handled. While it is seen that members from different cultures research ethics
differently, having a more flexible ethical approach which encompasses a flexible methodological
approach is called for. This certainly requires the identification and appreciation of cultural value,
norms and behaviours. Participant feedback is a crucial element of the research design as it will
353
help in identifying areas of improvements in the study. Future research should focus on
highlighting more cross-cultural challenges that researchers could encounter and recommend ways
to overcome them. Therefore, keeping all the above points in mind the present research was
conducted and the below section outlines an overview of the studies in this thesis.